Current AI rankings

Best AI Models & Tools

Find the best AI for what you want to create.

The one to beat

Best Coding Overall

Winner · #1

Claude Code + Opus 5.5 High

Anthropic / Claude Code

Demanding implementation and everyday codebase changes

Model API$4 in $20 out per 1M tokens
* Early evidence — position may changeTested configuration
Why it wins & tradeoffs

An early lead at High effort in a six-project comparison, backed by benchmarks and daily-use reports. Serious failures keep the recommendation provisional.

Price & access
Claude paid plan or API billing; select Opus 5.5 High. Cost per finished task varies by workload.
Watch out for
Keep human review and tests in the loop. Reports include broken workflows; long-running unattended reliability is not established.
Content restrictions
Moderate
Evidence
A six-project test and Artificial Analysis favor Opus over Astra at High effort. Daily-use reports include both useful patches and serious failures; broader replication is missing.
System version
September 2026 Claude Code; exact build undisclosed in external test
Tested access
Claude paid plan or API billing; High effort; default safeguards
Sources and observations (4)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot: Claude Code with Opus 5.5 High leads the six-project aggregate, ahead of Codex Astra High, with lower measured average API cost and task time. Perfect published behavioral subtotal; strong quality grading. Actual evaluation dated September 23. Limited projects and repetitions, automated quality judging, harness build undisclosed; publisher-specific sponsorship unknown. This directly tests High in Claude Code; not a debugging-specialty result.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 26 comparison: Opus 5.5 adaptive High with default fallback leads Astra High on Terminal-Bench 4.0 and SciCode, but Astra xhigh remains ahead of Opus High on Terminal-Bench. AA's model evaluation is contextual for Claude Code, not a Claude Code harness score. Intelligence-index cost per task is slightly higher than Astra High despite lower token prices; do not promise universally cheaper completion.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Claude Code + Opus 5.5 High: creator preferred its outputs on all three builds for detail, prompt fidelity and gameplay. Supports creative coding quality; does not establish superiority on repository repair or every coding task. Creator still prefers Astra for everyday refactors/architecture. Scope: creative coding support within the broader coding rubric; repository performance remains unmeasured here.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 Vibe Check: original Claude Code/app tests support daily implementation and concise patches, with some builders switching from Fable/Codex. Contrary evidence includes broken core app flows, uncalled required service and destructive verification behavior. Examples use mixed efforts (including xhigh), so they are contextual for High, not matched High victories. Every received Anthropic pre-launch access; editorial independence asserted by Every, no independent-replication claim.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

New evidence since this ranking (1)

These findings have not yet changed the ranking. Several articles from one publisher count as one source.

  • AI Coding Daily — original coding benchmarkSource record · Reviewed Oct 3, 2026

    October 3 table: Opus 5.5 High / Claude Code; $0.79, 3m10s, 67.41/70. Added bug-finding task changes denominator from 60; not a longitudinal improvement. Exact build and added-task rubric incompletely disclosed. One publisher; no standalone debugging qualification. Preserve prior observations.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 23, 2026Provider, pricing & access ↗
2

Codex + GPT-6 Astra

OpenAI / Codex

Codebase and terminal work with adjustable reasoning effort

* Early evidence — position may changeTested configuration
When to choose Codex + GPT-6 Astra

A close alternative that completed the test's required behaviors but trailed Opus at High effort. Extra-high terminal results are a separate strength.

Model API$10 in $50 out per 1M tokens
Price & access
Codex, ChatGPT Work or API; access and cost depend on plan and effort.
Watch out for
Opus leads the reviewed six-project test at High effort. Astra's stronger terminal results use a different effort setting.
Content restrictions
Low
Evidence
Opus leads the reviewed High-versus-High coding comparisons. Astra leads separate terminal tests at extra-high effort; those settings are not interchangeable.
System version
September 2026 production harness
Tested access
Codex, ChatGPT Work, or API
Sources and observations (5)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot explicitly tests Codex CLI + GPT-6 Astra High and Medium separately. High trails Claude Code + Opus 5.5 High on aggregate quality; both clear the behavior tasks. High is slower and costlier on this suite. High evidence is a configuration-specific observation within the historically broad Astra entry, not evidence about xhigh/max settings.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 26: Astra High trails Opus 5.5 High on terminal and scientific coding, while Astra xhigh retains a terminal advantage over Opus High. AA evaluates a different harness from Codex; settings must stay explicit. Astra retains stronger AutomationBench results. The broad Astra configuration supports choosing xhigh for hard terminal work, not claiming its High setting wins.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Codex + GPT-6 Astra, explicitly High in this comparison: Opus won creator preference on all three creative builds; Astra was faster in reported runs and cheapest on two. Creator still favors Astra for everyday refactors/architecture, a reported workflow preference rather than a measured debugging test. Do not transfer to xhigh or other efforts. Scope: creative coding support within the broader coding rubric; repository performance remains unmeasured here.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 review retains Astra for time-bounded work and reports higher raw throughput in one coding experiment, while Opus met more latency budgets. Scope is one experiment with mixed settings, not an Astra High benchmark. September 3 production-rewrite benchmark remains separate and used Astra extra-high.

  • Nate Herk — GPT-6 Astra vs Fable 5.1 practical comparisonSource page · Observed Sep 6, 2026

    Both configurations completed substantial practical work. The comparison favored Codex + Astra more often for day-to-day execution while preserving Claude Code + Fable 5.1 as stronger on selected tasks.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
3

Claude Code + Fable 5.1

Anthropic / Claude Code

Difficult implementation and careful changes to existing code

* Early evidence — position may changeTested configuration
When to choose Claude Code + Fable 5.1

A strong choice for difficult codebase problems, though newer comparisons favor other models. Tests use different effort settings.

Model API$10 in $50 out per 1M tokens
Price & access
Claude subscription or API billing; heavier usage can be costly.
Watch out for
Newer models lead some implementation tests and may cost less per task.
Content restrictions
Moderate
Evidence
Every still prefers Fable for some hard tasks. Other comparisons use Medium or different effort settings, so they do not establish a uniform defeat.
System version
September 2026 production release
Tested access
Claude Max or API billing
Sources and observations (5)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot: Claude Code + Fable 5.1 Medium trails Opus 5.5 High and Astra High on the six-project aggregate. The Fable row is Medium: it does not measure Fable High/Max. Retain that scope and use separate evidence for hard-problem frontier capability.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 22 original Opus 5.5 analysis: Opus max surpasses Fable 5.1's prior SciCode result and overall index, but this max comparison is contextual and cannot be relabeled Opus High versus Fable High. Fable remains a frontier coding model with premium pricing. Retain the older direct Fable configuration observations alongside this version-specific challenge.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Claude Code + Fable 5.1, explicitly High in this comparison: Opus won creator preference across all three creative builds. Fable showed speed but missed some requirements. These are task-level observations, not a universal third-place claim or evidence for other effort settings. Scope: creative coding support within the broader coding rubric; repository performance remains unmeasured here.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 review: some daily coding moves to Opus for value, but two testers still prefer Fable 5.1 on the hardest coding problems. Ruby comparison used Fable High versus Opus xhigh; do not treat effort as matched. This supports retaining Fable as a demanding-task alternative. Anthropic supplied pre-launch access to Opus; keep that relationship disclosed.

  • Nate Herk — GPT-6 Astra vs Fable 5.1 practical comparisonSource page · Observed Sep 6, 2026

    Both configurations completed substantial practical work. The comparison favored Codex + Astra more often for day-to-day execution while preserving Claude Code + Fable 5.1 as stronger on selected tasks.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
What we judge

Our strongest all-around choice for planning, building, testing, and maintaining code. We weigh quality, reliability, workflow, and cost together.

CorrectnessRepository understandingTool useMaintainabilityWorkflow reliability
About this provisional ranking

Based on external tests and hands-on comparisons. Results depend on the model, reasoning effort and coding tools. More testing may change the order; reviews are due every two days.

Models, coding tools, and builders: what’s the difference?
The model
Astra is the AI doing the reasoning. The tools it can use depend on where you access it.
The coding tool
Codex gives a model tools to work on project files, run tests, and fix problems. “Codex + Astra” names the combination being compared.
The builder
A platform such as Lovable brings building, databases, and publishing into one service. We judge the whole experience, including how easily you can maintain the result.

And ChatGPT Work? It’s a broader workspace for completing tasks; Codex provides the coding capabilities. Work and Astra aren’t competing models. The model, available tools, and setup together affect the result.

A tool can appear in more than one category. A website primarily presents information; a web app lets people do things with data. Some projects do both.

Advanced comparisonsAutonomous coding · Writing code · Fixing & improving code

Why doesn’t the overall winner win every category?

Overall balances several kinds of work, reliability, workflow, and cost. A specialist can produce a more refined implementation or a better fix while another tool offers the strongest all-around combination. Builder rankings also judge the finished product and publishing experience, so their winners can differ.

How we weigh the evidence →
Winner · #1

Codex + GPT-6 Astra

OpenAI / Codex

Codebase and terminal work with adjustable reasoning effort

Model API$10 in $50 out per 1M tokens
* Early evidence — position may changeTested configuration
Why it wins & tradeoffs

Leads for broader terminal and production work, including extra-high effort. Opus wins the reviewed coding comparisons when both use High effort.

Price & access
Codex, ChatGPT Work or API; access and cost depend on plan and effort.
Watch out for
Opus leads the reviewed six-project test at High effort. Astra's stronger terminal results use a different effort setting.
Content restrictions
Low
Evidence
Opus leads the reviewed High-versus-High coding comparisons. Astra leads separate terminal tests at extra-high effort; those settings are not interchangeable.
System version
September 2026 production harness
Tested access
Codex, ChatGPT Work, or API
Sources and observations (5)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot explicitly tests Codex CLI + GPT-6 Astra High and Medium separately. High trails Claude Code + Opus 5.5 High on aggregate quality; both clear the behavior tasks. High is slower and costlier on this suite. High evidence is a configuration-specific observation within the historically broad Astra entry, not evidence about xhigh/max settings.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 26: Astra High trails Opus 5.5 High on terminal and scientific coding, while Astra xhigh retains a terminal advantage over Opus High. AA evaluates a different harness from Codex; settings must stay explicit. Astra retains stronger AutomationBench results. The broad Astra configuration supports choosing xhigh for hard terminal work, not claiming its High setting wins.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Codex + GPT-6 Astra, explicitly High in this comparison: Opus won creator preference on all three creative builds; Astra was faster in reported runs and cheapest on two. Creator still favors Astra for everyday refactors/architecture, a reported workflow preference rather than a measured debugging test. Do not transfer to xhigh or other efforts. Scope: extended autonomous tool-using builds; not controlled repository repair. Deployment reminder/intervention limits one-shot claims.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 review retains Astra for time-bounded work and reports higher raw throughput in one coding experiment, while Opus met more latency budgets. Scope is one experiment with mixed settings, not an Astra High benchmark. September 3 production-rewrite benchmark remains separate and used Astra extra-high.

  • OpenAI Codex changelogSource page · Observed Sep 18, 2026

    Codex CLI 0.155.0 added live reasoning-status updates, agent-task archiving and worktree ownership, Touch ID verification for local MCP requests, recoverable daemon updates, and authentication recovery; 0.155.1 corrected reasoning-summary defaults for providers that reject them. These are material harness workflow and reliability changes, not direct model-quality evidence.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
2

Claude Code + Opus 5.5 High

Anthropic / Claude Code

Demanding implementation and everyday codebase changes

* Early evidence — position may changeTested configuration
When to choose Claude Code + Opus 5.5 High

Strong High-effort coding results, but broken workflows and a deleted verification file limit confidence in sustained unattended work.

Model API$4 in $20 out per 1M tokens
Price & access
Claude paid plan or API billing; select Opus 5.5 High. Cost per finished task varies by workload.
Watch out for
Keep human review and tests in the loop. Reports include broken workflows; long-running unattended reliability is not established.
Content restrictions
Moderate
Evidence
A six-project test and Artificial Analysis favor Opus over Astra at High effort. Daily-use reports include both useful patches and serious failures; broader replication is missing.
System version
September 2026 Claude Code; exact build undisclosed in external test
Tested access
Claude paid plan or API billing; High effort; default safeguards
Sources and observations (4)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot: Claude Code with Opus 5.5 High leads the six-project aggregate, ahead of Codex Astra High, with lower measured average API cost and task time. Perfect published behavioral subtotal; strong quality grading. Actual evaluation dated September 23. Limited projects and repetitions, automated quality judging, harness build undisclosed; publisher-specific sponsorship unknown. This directly tests High in Claude Code; not a debugging-specialty result.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 26 comparison: Opus 5.5 adaptive High with default fallback leads Astra High on Terminal-Bench 4.0 and SciCode, but Astra xhigh remains ahead of Opus High on Terminal-Bench. AA's model evaluation is contextual for Claude Code, not a Claude Code harness score. Intelligence-index cost per task is slightly higher than Astra High despite lower token prices; do not promise universally cheaper completion.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Claude Code + Opus 5.5 High: creator preferred its outputs on all three builds for detail, prompt fidelity and gameplay. Supports creative coding quality; does not establish superiority on repository repair or every coding task. Creator still prefers Astra for everyday refactors/architecture. Scope: extended autonomous tool-using builds; not controlled repository repair. Deployment reminder/intervention limits one-shot claims.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 Vibe Check: original Claude Code/app tests support daily implementation and concise patches, with some builders switching from Fable/Codex. Contrary evidence includes broken core app flows, uncalled required service and destructive verification behavior. Examples use mixed efforts (including xhigh), so they are contextual for High, not matched High victories. Every received Anthropic pre-launch access; editorial independence asserted by Every, no independent-replication claim.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 23, 2026Provider, pricing & access ↗
3

Claude Code + Fable 5.1

Anthropic / Claude Code

Difficult implementation and careful changes to existing code

* Early evidence — position may changeTested configuration
When to choose Claude Code + Fable 5.1

A strong choice for difficult codebase problems, though newer comparisons favor other models. Tests use different effort settings.

Model API$10 in $50 out per 1M tokens
Price & access
Claude subscription or API billing; heavier usage can be costly.
Watch out for
Newer models lead some implementation tests and may cost less per task.
Content restrictions
Moderate
Evidence
Every still prefers Fable for some hard tasks. Other comparisons use Medium or different effort settings, so they do not establish a uniform defeat.
System version
September 2026 production release
Tested access
Claude Max or API billing
Sources and observations (5)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot: Claude Code + Fable 5.1 Medium trails Opus 5.5 High and Astra High on the six-project aggregate. The Fable row is Medium: it does not measure Fable High/Max. Retain that scope and use separate evidence for hard-problem frontier capability.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 22 original Opus 5.5 analysis: Opus max surpasses Fable 5.1's prior SciCode result and overall index, but this max comparison is contextual and cannot be relabeled Opus High versus Fable High. Fable remains a frontier coding model with premium pricing. Retain the older direct Fable configuration observations alongside this version-specific challenge.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Claude Code + Fable 5.1, explicitly High in this comparison: Opus won creator preference across all three creative builds. Fable showed speed but missed some requirements. These are task-level observations, not a universal third-place claim or evidence for other effort settings. Scope: extended autonomous tool-using builds; not controlled repository repair. Deployment reminder/intervention limits one-shot claims.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 review: some daily coding moves to Opus for value, but two testers still prefer Fable 5.1 on the hardest coding problems. Ruby comparison used Fable High versus Opus xhigh; do not treat effort as matched. This supports retaining Fable as a demanding-task alternative. Anthropic supplied pre-launch access to Opus; keep that relationship disclosed.

  • Claude Code changelogSource page · Observed Sep 18, 2026

    Claude Code 2.1.274–2.1.276 added critical memory warnings, configurable MCP startup waits, synced claude.ai skills/plugins, marketplace plugin installation, improved queued-message control, and then repaired a proxy/gateway regression introduced in 2.1.275. These are material agent workflow and reliability changes, not direct task-quality benchmark evidence.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
What we judge

For delegating a multi-step development task: the agent plans, edits files, runs tests, and works through errors with limited supervision.

Task completionPlanningTool useReliability on long tasksRecovery from errors
About this provisional ranking

Based on external tests and hands-on comparisons. Results depend on the model, reasoning effort and coding tools. More testing may change the order; reviews are due every two days.

Winner · #1

Claude Code + Opus 5.5 High

Anthropic / Claude Code

Demanding implementation and everyday codebase changes

Model API$4 in $20 out per 1M tokens
* Early evidence — position may changeTested configuration
Why it wins & tradeoffs

An early lead at High effort in a six-project comparison, backed by benchmarks and daily-use reports. Serious failures keep the recommendation provisional.

Price & access
Claude paid plan or API billing; select Opus 5.5 High. Cost per finished task varies by workload.
Watch out for
Keep human review and tests in the loop. Reports include broken workflows; long-running unattended reliability is not established.
Content restrictions
Moderate
Evidence
A six-project test and Artificial Analysis favor Opus over Astra at High effort. Daily-use reports include both useful patches and serious failures; broader replication is missing.
System version
September 2026 Claude Code; exact build undisclosed in external test
Tested access
Claude paid plan or API billing; High effort; default safeguards
Sources and observations (4)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot: Claude Code with Opus 5.5 High leads the six-project aggregate, ahead of Codex Astra High, with lower measured average API cost and task time. Perfect published behavioral subtotal; strong quality grading. Actual evaluation dated September 23. Limited projects and repetitions, automated quality judging, harness build undisclosed; publisher-specific sponsorship unknown. This directly tests High in Claude Code; not a debugging-specialty result.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 26 comparison: Opus 5.5 adaptive High with default fallback leads Astra High on Terminal-Bench 4.0 and SciCode, but Astra xhigh remains ahead of Opus High on Terminal-Bench. AA's model evaluation is contextual for Claude Code, not a Claude Code harness score. Intelligence-index cost per task is slightly higher than Astra High despite lower token prices; do not promise universally cheaper completion.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Claude Code + Opus 5.5 High: creator preferred its outputs on all three builds for detail, prompt fidelity and gameplay. Supports creative coding quality; does not establish superiority on repository repair or every coding task. Creator still prefers Astra for everyday refactors/architecture. Scope: generated code/output fidelity across three tasks; auxiliary image tools must not be credited to native model generation.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 Vibe Check: original Claude Code/app tests support daily implementation and concise patches, with some builders switching from Fable/Codex. Contrary evidence includes broken core app flows, uncalled required service and destructive verification behavior. Examples use mixed efforts (including xhigh), so they are contextual for High, not matched High victories. Every received Anthropic pre-launch access; editorial independence asserted by Every, no independent-replication claim.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 23, 2026Provider, pricing & access ↗
2

Codex + GPT-6 Astra

OpenAI / Codex

Codebase and terminal work with adjustable reasoning effort

* Early evidence — position may changeTested configuration
When to choose Codex + GPT-6 Astra

A close alternative that completed the test's required behaviors but trailed Opus at High effort. Extra-high terminal results are a separate strength.

Model API$10 in $50 out per 1M tokens
Price & access
Codex, ChatGPT Work or API; access and cost depend on plan and effort.
Watch out for
Opus leads the reviewed six-project test at High effort. Astra's stronger terminal results use a different effort setting.
Content restrictions
Low
Evidence
Opus leads the reviewed High-versus-High coding comparisons. Astra leads separate terminal tests at extra-high effort; those settings are not interchangeable.
System version
September 2026 production harness
Tested access
Codex, ChatGPT Work, or API
Sources and observations (5)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot explicitly tests Codex CLI + GPT-6 Astra High and Medium separately. High trails Claude Code + Opus 5.5 High on aggregate quality; both clear the behavior tasks. High is slower and costlier on this suite. High evidence is a configuration-specific observation within the historically broad Astra entry, not evidence about xhigh/max settings.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 26: Astra High trails Opus 5.5 High on terminal and scientific coding, while Astra xhigh retains a terminal advantage over Opus High. AA evaluates a different harness from Codex; settings must stay explicit. Astra retains stronger AutomationBench results. The broad Astra configuration supports choosing xhigh for hard terminal work, not claiming its High setting wins.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Codex + GPT-6 Astra, explicitly High in this comparison: Opus won creator preference on all three creative builds; Astra was faster in reported runs and cheapest on two. Creator still favors Astra for everyday refactors/architecture, a reported workflow preference rather than a measured debugging test. Do not transfer to xhigh or other efforts. Scope: generated code/output fidelity across three tasks; auxiliary image tools must not be credited to native model generation.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 review retains Astra for time-bounded work and reports higher raw throughput in one coding experiment, while Opus met more latency budgets. Scope is one experiment with mixed settings, not an Astra High benchmark. September 3 production-rewrite benchmark remains separate and used Astra extra-high.

  • Nate Herk — GPT-6 Astra vs Fable 5.1 practical comparisonSource page · Observed Sep 6, 2026

    The software-build task compared the exact Codex + Astra and Claude Code + Fable 5.1 configurations rather than attributing results to naked models.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
3

Claude Code + Fable 5.1

Anthropic / Claude Code

Difficult implementation and careful changes to existing code

* Early evidence — position may changeTested configuration
When to choose Claude Code + Fable 5.1

A strong choice for difficult codebase problems, though newer comparisons favor other models. Tests use different effort settings.

Model API$10 in $50 out per 1M tokens
Price & access
Claude subscription or API billing; heavier usage can be costly.
Watch out for
Newer models lead some implementation tests and may cost less per task.
Content restrictions
Moderate
Evidence
Every still prefers Fable for some hard tasks. Other comparisons use Medium or different effort settings, so they do not establish a uniform defeat.
System version
September 2026 production release
Tested access
Claude Max or API billing
Sources and observations (5)
  • AI Coding Daily — original coding benchmarkSource record · Observed Sep 26, 2026

    September 25 leaderboard snapshot: Claude Code + Fable 5.1 Medium trails Opus 5.5 High and Astra High on the six-project aggregate. The Fable row is Medium: it does not measure Fable High/Max. Retain that scope and use separate evidence for hard-problem frontier capability.

  • Artificial AnalysisSource record · Observed Sep 26, 2026

    AA September 22 original Opus 5.5 analysis: Opus max surpasses Fable 5.1's prior SciCode result and overall index, but this max comparison is contextual and cannot be relabeled Opus High versus Fable High. Fable remains a frontier coding model with premium pricing. Retain the older direct Fable configuration observations alongside this version-specific challenge.

  • Pat Simmons / AI for Mortals — practical coding comparisonsSource record · Observed Sep 23, 2026

    Original comparison VxzdNX6mNSQ and companion AI for Mortals results article; tests September 23, 2026. Same prompts, High effort on all models in Claude Code/Codex, three creative builds (code-only illustration, launch site/film, 3D game). Exact harness builds undisclosed; shared gpt-image-2 where allowed. Single creator's practical comparison; not a BMN reproduction, objective repository benchmark, or separate publisher from the article. Claude Code + Fable 5.1, explicitly High in this comparison: Opus won creator preference across all three creative builds. Fable showed speed but missed some requirements. These are task-level observations, not a universal third-place claim or evidence for other effort settings. Scope: generated code/output fidelity across three tasks; auxiliary image tools must not be credited to native model generation.

  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource record · Observed Sep 22, 2026

    Every September 22 Opus 5.5 review: some daily coding moves to Opus for value, but two testers still prefer Fable 5.1 on the hardest coding problems. Ruby comparison used Fable High versus Opus xhigh; do not treat effort as matched. This supports retaining Fable as a demanding-task alternative. Anthropic supplied pre-launch access to Opus; keep that relationship disclosed.

  • Nate Herk — GPT-6 Astra vs Fable 5.1 practical comparisonSource page · Observed Sep 6, 2026

    The software-build task compared the exact Codex + Astra and Claude Code + Fable 5.1 configurations rather than attributing results to naked models.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 26, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
What we judge

For turning clear instructions into correct, maintainable code. The focus is the implementation itself.

Functional correctnessFollowing instructionsCode qualityLanguage rangeEfficiency
About this provisional ranking

Based on external tests and hands-on comparisons. Results depend on the model, reasoning effort and coding tools. More testing may change the order; reviews are due every two days.

Winner · #1

Claude Code + Fable 5.1

Anthropic / Claude Code

Careful fixes and improvements to large codebases

Model API$10 in $50 out per 1M tokens
Weighted score: 92.2Tested configuration
Why it wins & tradeoffs

Leads for careful codebase judgment and disciplined changes.

Price & access
Claude Max or API billing; heavy usage can become expensive.
Watch out for
Can be slower and more expensive per completed task.
Content restrictions
Moderate
Evidence
Current independent benchmarks and practitioner testing support frontier quality and strong product judgment, with higher cost and slower task completion.
System version
September 2026 production release
Tested access
Claude Max or API billing
Sources and observations (1)

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 12, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
2

Codex + GPT-6 Astra

OpenAI / Codex

Fast codebase fixes, rewrites and terminal work

Weighted score: 90.6Tested configuration
When to choose Codex + GPT-6 Astra

Strong results on production rewrites and efficiency, but can make broader changes than wanted.

Model API$10 in $50 out per 1M tokens
Price & access
Codex, ChatGPT Work, or API; frontier access remains plan-dependent.
Watch out for
Review the scope of its changes; access also depends on your plan.
Content restrictions
Low
Evidence
Artificial Analysis, terminal tests, Every and practitioner comparisons support strong performance, speed and cost efficiency.
System version
September 2026 production harness
Tested access
Codex, ChatGPT Work, or API
Sources and observations (2)
  • Every — current coding-agent Vibe Checks and Senior Engineer BenchmarkSource page · Observed Sep 3, 2026

    Astra's 71/100 production-code rewrite materially exceeded both Fable 5.1 runs, but Every documented overbuilding and stopping-judgment weaknesses.

  • Artificial AnalysisReferenced source page · Observed Sep 12, 2026

    Referenced in the published evaluation or explanation. A specific source observation is not attached here; this link is context, not a substitute for a documented comparison.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 12, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
3

Claude Code + Opus 5

Anthropic / Claude Code

Root-cause diagnosis and careful code changes

Weighted score: 88.8Tested configuration

Supporting source links for this published explanation are still being reconciled.

When to choose Claude Code + Opus 5

A dependable third choice for diagnosing root causes and improving existing code.

Model API$5 in $25 out per 1M tokens
Price & access
Claude subscription or API billing.
Watch out for
Newer alternatives may perform better or cost less.
Content restrictions
Moderate
Evidence
Independent repository and coding-agent evidence supports mature, dependable engineering performance.
System version
September 2026 production release
Tested access
Claude Max or API billing
Ranking last verified: Sep 12, 2026 · Configuration tested: Sep 7, 2026Provider, pricing & access ↗
Also excellent

Cursor + Fable 5.1

Cursor / Anthropic

A close alternative to third place (0.2 weighted points apart). This gap does not establish a meaningful quality difference.

Model API$10 in $50 out per 1M tokens
Weighted score: 88.6
What we judge

For finding the cause of a bug or improving an existing project without breaking what already works.

DiagnosisFix correctnessRegression avoidanceRefactoring qualityExplanation

Website & app builders

What makes a good builder? →
Practitioner reviews

Best Website Builder

Winner · #1

Codex + GPT-6 Astra

OpenAI / Codex

Distinctive marketing sites with code ownership

Model API$10 in $50 out per 1M tokens
Weighted score: 92.3Provisional: evidence still developingTested configuration
Why it wins & tradeoffs

Our provisional design leader for polished custom sites, with strong layouts, motion and brand fit.

Price & access
Codex, ChatGPT Work, and API; API starts at $10/M input and $50/M output
Watch out for
Maintaining the site may require coding help; this is not a visual editor.
Content restrictions
Low
Evidence
Nate Herk's reference-site test and independent design tests support the lead. More practitioner comparisons are needed.
System version
September 2026 production harness
Tested access
Codex, ChatGPT Work, or API
Sources and observations (1)

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 10, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
2

Framer

Framer

Marketing sites and portfolios

Weighted score: 91.0Provisional: evidence still developingTested configurationPlatform-selected AI

Supporting source links for this published explanation are still being reconciled.

When to choose Framer

A provisional choice for polished sites with a visual editor and responsive design controls.

Price & access
Free entry tier with paid hosting and collaboration plans
Watch out for
Limited fit for complex app backends or unrestricted code ownership.
Content restrictions
Low
Evidence
Specialist reviews and verified features support a top-three place. More practitioner comparisons are needed.
System version
September 2026 production release
Tested access
Current public paid plan
AI model
Selected automatically by Framer. The exact model—or combination of models—may change as the service evolves.
Ranking last verified: Sep 10, 2026 · Configuration tested: Sep 12, 2026Provider, pricing & access ↗
3

Claude Code + Fable 5.1

Anthropic / Claude Code

Custom sites built through several rounds of feedback

Weighted score: 90.5Provisional: evidence still developingTested configuration
When to choose Claude Code + Fable 5.1

Produces polished custom sites after revision; needed more rounds than Astra in the reviewed comparison.

Model API$10 in $50 out per 1M tokens
Price & access
Claude Max or API billing
Watch out for
Expect revisions rather than a polished first attempt. There is no visual no-code editor.
Content restrictions
Moderate
Evidence
Nate Herk tested Claude Code with Fable 5.1 on the same reference-site task. More independent comparisons are needed.
System version
September 2026 production release
Tested access
Claude Max or API billing
Sources and observations (1)

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 10, 2026 · Configuration tested: Sep 6, 2026Provider, pricing & access ↗
What we judge

Create a portfolio, business homepage, or publication. We look for polished design, easy editing, and a reliable way to publish.

Practitioner reviewsFinished websitesOwnership & publishingEditing & recoveryEase & running costs
Practitioner reviews

Best Web App Builder

Winner · #1

Lovable

Lovable

Complete web-app prototypes with sign-in, data and GitHub export

Weighted score: 97.0Tested configurationPlatform-selected AI
Why it wins & tradeoffs

Three practitioner reviews put Lovable first, with evidence of working apps and code ownership.

Watch out for
Limited framework choice and fine-grained manual control.
Content restrictions
Low
Evidence
Three reviews rank it first; two are independent, and one shows a working, deployed project tracker.
System version
September 2026 production release
Tested access
Current public paid plan
AI model
Selected automatically by Lovable. The exact model—or combination of models—may change as the service evolves.
Sources and observations (3)
  • LOW/CODE Agency — 2026 no-code builder comparisonSource page · Observed Sep 3, 2026

    LOW/CODE's explicitly numbered vibe-coding section places Lovable first and calls it the best AI app builder for nontechnical founders, based on recent internal testing across real workflows.

  • Vibe Coding Academy — 2026 AI app builder guideSource page · Observed May 1, 2026

    Vibe Coding Academy names Lovable the strongest all-around AI web app builder and demonstrates a deployed project tracker with authentication, database-backed projects and tasks, and iterative dashboard changes.

  • Till Freitag — Base44 vs Lovable agency comparisonSource page · Observed Mar 17, 2026

    After using both builders on several agency projects and comparing a CRM dashboard workflow, Till Freitag prefers Lovable for production and client work because of code and backend portability; its Lovable services create a material commercial-interest discount.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 12, 2026 · Configuration tested: Sep 12, 2026Provider, pricing & access ↗
2

Bolt

StackBlitz

Quick web-app prototypes with code ownership

Weighted score: 49.3Tested configurationPlatform-selected AI
When to choose Bolt

A strong runner-up for fast revisions and exportable code.

Watch out for
Setting up the backend may take more work than with an all-in-one builder.
Content restrictions
Low
Evidence
LOW/CODE ranks Bolt second; provider documentation supports code ownership and publishing.
System version
September 2026 production release
Tested access
Current public paid plan
AI model
Selected automatically by Bolt. The exact model—or combination of models—may change as the service evolves.
Sources and observations (1)

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 12, 2026 · Configuration tested: Sep 12, 2026Provider, pricing & access ↗
3

Base44

Wix / Base44

Internal tools and data-heavy business apps

Weighted score: 37.8Tested configurationPlatform-selected AI
When to choose Base44

An approachable choice for internal tools, with less freedom to move the backend elsewhere.

Watch out for
Moving the backend to another platform may be difficult.
Content restrictions
Moderate
Evidence
LOW/CODE ranks Base44 third. Practitioner reviews praise ease of use but flag portability limits.
System version
September 2026 production release
Tested access
Current public paid plan
AI model
Selected automatically by Base44. The exact model—or combination of models—may change as the service evolves.
Sources and observations (1)

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 12, 2026 · Configuration tested: Sep 12, 2026Provider, pricing & access ↗
What we judge

Build something people use in a browser, such as a booking system, CRM, or writing tool—with accounts, saved data, and working features.

Practitioner reviewsWorking productsOwnership & publishingEase & revisionRecovery & running costs
Practitioner reviews

Best Mobile App Builder

Winner · #1

FlutterFlow

FlutterFlow

iOS and Android apps intended for the app stores

Weighted score: 89.7Provisional: evidence still developingTested configurationPlatform-selected AI
Why it wins & tradeoffs

Our provisional first choice for visual app building, with testing, code export and app-store publishing.

Price & access
Paid FlutterFlow plan required for key export and deployment workflows
Watch out for
You will need to learn the visual editor; chat alone is not enough.
Content restrictions
Low
Evidence
Publishing tools are verified. More independent comparisons building the same app are needed.
System version
September 2026 production release
Tested access
Current public paid plan
AI model
Selected automatically by FlutterFlow. The exact model—or combination of models—may change as the service evolves.
Sources and observations (3)
  • LOW/CODE Agency — 2026 no-code builder comparisonSource page · Observed Sep 3, 2026

    The agency identifies FlutterFlow as its strongest fit for cross-platform native apps needing source ownership, based on recent hands-on evaluation and production experience. The article does not publish a category-specific ranked top-three ballot.

  • The App Builder Report — Rork vs FlutterFlowSource page · Observed Aug 14, 2026

    The comparison explicitly judges FlutterFlow more capable than Rork for repeat building, custom logic, code ownership, and mature store deployment. FlutterFlow was assessed from documentation and independent coverage rather than the author's direct use; the publisher's Rork affiliate relationship discounts the verdict.

  • Unite.AI — Superapp hands-on reviewSource page · Observed Jul 31, 2026

    In the same hands-on review, FlutterFlow is characterized as the more mature choice for cross-platform control and customization. The comparison is useful but affiliate-discounted and is not a complete mobile top-three ballot.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 10, 2026 · Configuration tested: Sep 12, 2026Provider, pricing & access ↗
2

Draftbit

Draftbit

Visual mobile development with React Native code export

Weighted score: 85.3Provisional: evidence still developingTested configurationPlatform-selected AI
When to choose Draftbit

A provisional runner-up for visual editing with React Native, Expo and exportable code.

Price & access
Paid plans unlock production collaboration and export workflows
Watch out for
Publishing still takes setup; the no-code community is smaller than some alternatives.
Content restrictions
Low
Evidence
Verified tools and mobile-builder comparisons support inclusion. More comparative testing is needed.
System version
September 2026 production release
Tested access
Current public paid plan
AI model
Selected automatically by Draftbit. The exact model—or combination of models—may change as the service evolves.
Sources and observations (1)
  • LOW/CODE Agency — 2026 no-code builder comparisonSource page · Observed Sep 3, 2026

    The agency identifies Draftbit as a strong React Native option for semi-technical teams prioritizing source ownership. The article does not publish a category-specific ranked top-three ballot.

Review dates may differ from publication dates. Some links lead to a source page rather than an individual test.

Ranking last verified: Sep 10, 2026 · Configuration tested: Sep 12, 2026Provider, pricing & access ↗
3

Replit Agent

Replit

One workspace for coding, data and deployment

Weighted score: 85.2Provisional: evidence still developingTested configurationPlatform-selected AI

Supporting source links for this published explanation are still being reconciled.

When to choose Replit Agent

A provisional choice for building web and mobile apps in one workspace.

Price & access
Replit Core or Teams with usage-based agent costs
Watch out for
Agent usage costs can be hard to predict, and initial designs may need polishing.
Content restrictions
Low
Evidence
Verified features and builder comparisons support its range. Independent practitioner agreement is still incomplete.
System version
September 2026 production release
Tested access
Replit Core or Teams
AI model
Selected automatically by Replit Agent. The exact model—or combination of models—may change as the service evolves.
Ranking last verified: Sep 10, 2026 · Configuration tested: Sep 7, 2026Provider, pricing & access ↗
What we judge

Create an installable iPhone or Android app. Device testing, code ownership, and a practical path to app stores matter here.

Practitioner reviewsWorking mobile appsCode ownershipApp-store publishingEase, recovery & value

Small score gaps are close calls. Media rankings are editorial. How we compare →

Know when the recommendations change.

Meaningful AI updates, plus On the Horizon: new releases we’re watching. Tuesdays at 9 a.m. Pacific, when there’s something worth sharing.

API pricing

Leading text & coding models. US dollars per million tokens.

Prices checked Sep 26, 2026

Category winners + On the Horizon

ModelInput 0 — $10Output 0 — $50
Input$4
Output$20
Input$10
Output$50
Input$10
Output$50

Each column has its own scale. Standard uncached input and output rates; short-context pricing where applicable.

Input is what you send; output is what the model generates, including billable thinking. More thinking can mean more tokens at the same rate. These are API costs, not monthly subscriptions or a price per completed task.

Sources, thinking levels & pricing details

Compare standard paid API access. Caching, batch discounts, fast service, regional routing, tool fees and taxes can change the bill. Coding tools may add their own charges. Low, medium and high thinking are provider-specific settings, not comparable amounts of reasoning.

Gemini 3.8 Flash

Paid standard introductory rate through December 31, 2026; scheduled to become $1.50 input / $7.50 output on January 1, 2027. Output includes thinking tokens.

Provider pricing ↗Checked 2026-09-26

GPT-6 Sol

Standard rate for prompts up to 272K input tokens. Above 272K, the full request costs 2× the input rate and 1.5× the output rate. Low, medium and high reasoning share these token rates; extra reasoning adds billable output tokens.

Provider pricing ↗Checked 2026-09-26

Claude Opus 5.5

Global standard rate across the supported context window. Low, medium and high effort share these token rates; thinking adds billable output. Fast mode costs extra.

Provider pricing ↗Checked 2026-09-26

GPT-5.6 Sol

Promotional standard short-context rate, available at least through November 21, 2026. Above 272K input tokens, the full request costs 2× input and 1.5× output. Thinking adds billable output tokens.

Provider pricing ↗Checked 2026-09-26

Claude Opus 5

Global standard rate across the supported context window. Thinking is included in billed output; fast mode costs extra.

Provider pricing ↗Checked 2026-09-26

Claude Fable 5.1

Global standard rate across the supported context window. Thinking is included in billed output; effort changes token use, not this rate.

Provider pricing ↗Checked 2026-09-26

GPT-6 Astra

Standard rate for prompts up to 272K input tokens. Above 272K, the full request costs 2× the input rate and 1.5× the output rate. Low, medium and high reasoning share these token rates; extra reasoning adds billable output tokens.

Provider pricing ↗Checked 2026-09-26

Thinking billing: OpenAI · Anthropic · Google

On the Horizon

New models still being assessed. Each status shows what needs to happen next; completed reviews have their own record.

Completed reviews & decisions →

OpenAIGathering comparison evidence

GPT-6.2 Astra

Announced Nov 17, 2026

GPT-6.2 Astra is available in Work, Codex and the API. OpenAI attributes improved coding and professional workflows to this release; independent category assessment is pending.

Next step & review dates

Obtain exact-configuration independent hands-on comparisons alongside current benchmarks. Writing screens require category-specific external qualification and budget capacity; no paid screen authorized by registration.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

OpenAIGathering comparison evidence

GPT-6.2 Sol

Announced Nov 17, 2026

GPT-6.2 Sol is available in Work, Codex and the API. OpenAI attributes improved coding and professional workflows to this release; independent category assessment is pending.

Next step & review dates

Obtain exact-configuration independent hands-on comparisons alongside current benchmarks. Writing screens require category-specific external qualification and budget capacity; no paid screen authorized by registration.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MicrosoftGathering comparison evidence

MAI-Code-2

Announced Nov 17, 2026

Microsoft confirms public GitHub Copilot availability for MAI-Code-2. The registered coding configuration uses Copilot CLI Max; build is not disclosed in the launch evaluation.

Next step & review dates

Obtain independent original repository comparisons on Copilot CLI Max with build, attempts, interventions and complete costs. Microsoft MSBench comparisons are vendor claims and promotional pricing is not stable per-task cost.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MicrosoftGathering comparison evidence

MAI-Code-2-Flash

Announced Nov 17, 2026

Microsoft confirms GitHub Copilot availability and limited Microsoft Foundry preview for MAI-Code-2-Flash. Copilot CLI Max is registered distinctly from other routes.

Next step & review dates

Seek independent matched Copilot CLI Max comparisons; do not transfer Code2 or Code1.1Flash scores. Vendor frontend preference study and discounted prices alone do not establish comparative coding reliability.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MicrosoftGathering comparison evidence

MAI-Image-2.7

Announced Nov 17, 2026

Microsoft released MAI-Image-2.7 in Microsoft Foundry public preview on November17. This release is tracked distinctly from the October preview. Microsoft advertises generation and editing with transparency support.

Next step & review dates

Match exact November17 output to independent repeated photo, typography, artistry and same-source editing comparisons. Do not transfer October preview rankings or vendor claims. Separate standard and Flash routes.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MicrosoftGathering comparison evidence

MAI-Image-2.7-Flash

Announced Nov 17, 2026

Microsoft released the Flash variant of MAI-Image-2.7 in Foundry public preview on November17. Microsoft advertises a lower-cost faster option, also used in Canvas.

Next step & review dates

Seek original same-input standard-versus-Flash and incumbent comparisons with native resolution, repeat attempts, latency, failure cases and financial disclosures; no ranking inference from release pricing.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

AlibabaGathering comparison evidence

Qwen4-Max

Announced Nov 16, 2026

Alibaba released production Qwen4-Max for text output and multimodal understanding. The qwen4-max API defaults to High effort; it is distinct from qwen4-max-preview.

Next step & review dates

Production model docs recovered after press-page access failed. Atomic Agent's three runs on one59-test repo and CodeRouter's175Python tasks are narrow and use different harnesses/retry caps. Replicate production High on multiple repositories; seek approved production-specific writing evidence before paid work.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MeituanGathering comparison evidence

LongCat-Video-Avatar 1.6

Announced Nov 15, 2026

Published weights and CUDA commands support audio-driven text/image-conditioned single- and multi-character video. Seek matched lip sync, role separation, identity drift and elapsed compute versus another audio-driven system.

Next step & review dates

Published weights and CUDA commands support audio-driven text/image-conditioned single- and multi-character video. Seek matched lip sync, role separation, identity drift and elapsed compute versus another audio-driven system.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

inclusionAIGathering comparison evidence

Ling 3.2 Flash

Announced Nov 13, 2026

Ant Ling documents its native Nov5 launch and temporary free access. OpenRouter lists a separate Nov13 rate-limited Ant Digital route with 1M context and text/image input. Original browser coding trial is scoped to its own harness; no writing quality conclusion.

Next step & review dates

Ant’s native launch was Nov5; OpenRouter’s distinct free route lists Nov13. Original four HTML tasks on the exact free route (thinking off) returned a turret error and an upstream timeout. Seek a dated fiction result or original screenplay comparison, separately; this coding task qualifies neither.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Alibaba QwenGathering comparison evidence

Qwen4-Max-Preview

Announced Nov 11, 2026

Alibaba released a fixed Qwen4-Max-Preview checkpoint through Token Plan, with default xhigh thinking and a1M context. The preview window runs November11–25; dedicated Token Plan credentials are required.

Next step & review dates

The production qwen4-max launched Nov16 as a separate endpoint. Preview Token Plan expires Nov25 (UTC+8). Do not transfer production Atomic Agent 0.7.0 or CodeRouter results; track retirement or an exact preview comparison.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

BaiduGathering comparison evidence

ERNIE-Light-DT-1.0

Announced Nov 10, 2026

Distinct baidu/ernie-light-dt-1.0-intl endpoint served by Baidu Qianfan, separate from ERNIE 5.5 Flash Preview. Seek dated exact-route writing-specific observations; no paid screen until qualified.

Next step & review dates

Distinct baidu/ernie-light-dt-1.0-intl endpoint served by Baidu Qianfan, separate from ERNIE 5.5 Flash Preview. Seek dated exact-route writing-specific observations; no paid screen until qualified.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

AnthropicAwaiting public access

Claude Mythos 5.2

Announced Nov 9, 2026

Anthropic announced Mythos 5.2 for verified organizations. Access is restricted.

Next step & review dates

Verify authorized access to an exact usable production configuration before evaluating it. Fable’s API availability and writing tests do not establish Mythos eligibility or quality.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

ByteDanceGathering comparison evidence

Seedream 5.5

Announced Nov 9, 2026

ByteDance introduced Seedream 5.5 for image generation and editing. Dreamina documents the Image 5.5 selector for eligible accounts; ByteDance reports native transparency and improved preservation across successive edits.

Next step & review dates

Saybrush Nov11 kie API 1.5K PNG chains show subject preservation in successive wardrobe and scene edits. One commercial host, selected examples, no same-input rivals. Compare native exact setting on repeated edits, alpha, small text and unedited regions against incumbents.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

OpenAIAwaiting public access

GPT-6 Cyber

Announced Nov 5, 2026

OpenAI’s November 5 security model is initially limited to selected approved Daybreak Red customers. Vendor cyber evaluations are not a general coding ranking.

Next step & review dates

Confirm eligible access and exact harness/effort before registering a rankable coding configuration. No general writing screen warranted.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

GoogleGathering comparison evidence

Gemini Omni 1.2 Flash

Announced Nov 5, 2026

Google released Omni 1.2 Flash with native 15-second generation and editing at $0.07/second for 720p. Arena's preliminary text-to-video board puts it first; this is not a BMN overall quality verdict.

Next step & review dates

CellCog's original11-task comparison shows promising lifelike dialogue but API restrictions changed four tasks and older competitor clips were reused; publisher sells competing generation. Audio-inclusive AA preferences are not picture-only quality. Obtain repeated same-footage edits with faces/background preserved and matched instructions.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Thinking Machines LabAwaiting testing access

Inkling 2 Small

Announced Nov 5, 2026

Thinking Machines' Nov12 update and public HF model upload confirm Inkling2Small weights are now available under Apache2.0, with additional hosting and runtime integrations. Vendor benchmarks remain claims; this access change does not release saved BMN writing holds.

Next step & review dates

Nov12 official update and HF upload verify Apache2 weights released; access metadata changed. Saved writing holds45/46 remain until a compliant unchanged-target exact route passes a documented availability decision. Then external category qualification and $5 reservations still apply. Coding comparisons of Tinker256K/OpenCode/other hosts remain distinct.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MeituanAwaiting testing access

LongCat-2.5

Announced Nov 2, 2026

Meituan's final LongCat-2.5 API is available, distinct from September's preview; MIT weights and native API documentation were verified November11. Vendor benchmarks use several harnesses and do not independently establish comparative quality.

Next step & review dates

Native API/docs and TOKRACE creative examples checked; mechanical/diagram errors and useful assets are task-specific, not matched ClaudeCode2.1.280/512K evidence. Obtain exact independent coding and professional-writing comparisons. Native BMN credential unverified; any future creative screen also requires external category qualification and budget eligibility.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

PikaGathering comparison evidence

Pika H3 Flex

Announced Oct 29, 2026

Pika released H3 Flex with native 480p/768p video and synchronized audio. Its hosted upscaler is separate.

Next step & review dates

Verify repeated native-resolution reference motion and changes to existing footage against Kling and Seedance; separate native detail from upscaling.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

SourcefulGathering comparison evidence

Riverflow 2.6 Pro High

Announced Oct 29, 2026

Sourceful2.6Pro high-effort image configuration verified November12. Delivered custom sizes may include resizing or upscaling; vendor GEditBench preference claims are not independent evidence.

Next step & review dates

Registered47946. Obtain original same-input photo, editing and design comparisons against current field with native-output inspection, effort, repeats, failures and commercial disclosures. No authorized paid internal media testing.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Particle AIGathering comparison evidence

Particle-1

Announced Oct 28, 2026

Vercel lists Particle-1, a coding-focused post-training of DeepSeek V4.1 Flash, on its AI Gateway. Vercel documents use through Codex and other coding clients.

Next step & review dates

Particle’s Oct 28 evaluation used Terminus 2 at Max, 16 runs per task and an assumed 95% cache hit rate. It does not evaluate Codex at the registered default. Next compare the same repositories, retries, correctness, total tokens and cost on that route.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Nanbeige LLM LabGathering comparison evidence

Nanbeige4.3-33B

Announced Oct 24, 2026

The lab released the 33B instruction checkpoint for local inference. The card distinguishes Base and FP8 and documents mini-SWE-agent at the highest thinking setting.

Next step & review dates

Local weights remain available; require category-specific external writing qualification, compliant unchanged-length serving route and conservative budget reservation before any new screen. Coding needs exact mini-SWE-agent2.4.6/preserved-thinking independent repository comparisons.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

NVIDIAAwaiting testing access

Nemotron 3.5 Bolt E4B

Announced Oct 23, 2026

Official Oct23 model card, actual Q4_K_M GGUF file and separate public NVIDIA NIM text endpoint verified Oct24. Local checkpoint documents64K context. Separate from Super VL and Bolt74B; vendor quality claims uncorroborated.

Next step & review dates

Preserve availability holds37/38 and saved work. Original NIM route documentation does not supply the missing NVIDIA credential or establish unchanged-target usability. Release only by documented availability decision, followed by category-specific writing qualification and budget gates. No quality inference.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

TopviewGathering comparison evidence

Topview T1

Announced Oct 21, 2026

Topview T1 is available through paid text, image and omni-reference video API routes. The provider offers 720p and 1080p output. AI Samson’s Oct 22 T1/Seedance 2.5 comparison is one creator’s evidence.

Next step & review dates

The Nov 4 seven-input case demonstrates reference-directed character replacement. Fourteen indexed comparisons are leads, not controlled trials. Obtain original repeat attempts on the same single still and multi-reference inputs against Seedance 2.5 and Kling; retain resolution and funding disclosures.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Black Forest LabsGathering comparison evidence

FLUX 3 Image Klein

Announced Oct 15, 2026

Black Forest Labs released FLUX 3 Image Klein on October15 with public API and license-gated downloadable weights. Distilled16-step and Base50-step CFG3.5 variants are registered separately. The5.6B label covers the image transformer, excluding text encoder and prompt upsampler. Capabilities and speed statements are vendor claims, not a BMN quality verdict.

Next step & review dates

An Oct 16–17 local RTX 5090 comparison gives distilled int8 Klein the edge on short lettering and layouts; Qwen handled long text and some relighting better. Prompt enhancement changed outcomes. Match these tasks on the native BFL API and Base BF16 before assigning either a local result.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

KreaGathering comparison evidence

Krea 3

Announced Oct 15, 2026

Krea announced Krea3 on October15. The native release describes image editing with reference inputs; partner fal also lists a text-to-image route. Native and partner modes, resolutions and Ultra/Fast variants require separate identification. Vendor speed and creative-quality claims are not a BMN verdict.

Next step & review dates

Use the 200-edit hosted blind study for editing tendencies only. Match native Krea 3 Edit versus Ultra/Fast routes and obtain independent evidence for generation and artistry.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

GitHubGathering comparison evidence

GitHub Copilot HydraFusion / Max — October update

Announced Oct 14, 2026

GitHub updated HydraFusion's runtime orchestration and introduced Max on October14. Both remain research previews across Copilot clients. Effort and model selection vary by stage; Max can use extra-high effort. GitHub's capability claims are not a BMN quality verdict.

Next step & review dates

Artificial Analysis tested generally available Fusion Max in Copilot CLI on October 28. Terminal tasks favored Fusion; the overall index and repository questions favored Claude Code Opus Max. Seek independent completed-project evidence with the allowed model pool and intervention history. No transfer to Standard, High or HydraFusion preview.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MetaGathering comparison evidence

Muse Wonder Full

Announced Oct 12, 2026

Meta released Muse Wonder Full on October 12 as a distinct model available to Muse Code subscribers. Official ID muse-wonder-full-1.0; 1M context. Meta documents Power Usage and Maximum subscription access. No public API route was verified. Performance statements are vendor claims, not a BMN quality verdict.

Next step & review dates

c0ze’s Oct17 13-task suite includes Muse Code xhigh; Contributor state not verified. VulcanBench tested Contributor mode and Pat tested goal mode with Contributor off. Obtain mode-matched tasks rather than combining the three.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Z.aiGathering comparison evidence

GLM-Legal

Announced Oct 12, 2026

Z.ai announced GLM-Legal on October 12, built on GLM-5.3-Flash for legal workflows. The vendor describes specialized training and invites applications for a beta, with AutoClaw integration coming later. Performance claims are vendor claims, not a BMN quality verdict.

Next step & review dates

An exact public legal route is now verified: z-ai/glm-5.3-flashx-legal. Verify correspondence to the original beta and compare source-bound legal drafting against incumbents; availability does not establish general business-writing quality.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

JetBrainsAwaiting testing access

Mellum2.1 Thinking

Announced Oct 8, 2026

JetBrains released Mellum2.1, including public Apache-2.0 Thinking weights with 12B total and 2.5B active parameters and 131072-token context. The official model card documents vLLM deployment and thinking settings. Coding improvements are vendor-run results, not a BMN verdict.

Next step & review dates

November11 JetBrains original Hugging Face card confirms BF16 Apache2 weights,131072context and vLLM qwen3/Hermes tooling; vendor Pi0.73.1/114Kcontext evaluation is not independent. No hosted inference provider listed. Next obtain an authorized BMN route or independent exact-runtime repository comparison.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

OpenAIGathering comparison evidence

GPT-6 Sol — October 2026 ChatGPT

Announced Oct 7, 2026

OpenAI released an October 7 GPT-6 Sol snapshot in ChatGPT alongside Intelligent UI. First-party deployment documentation explicitly distinguishes the October ChatGPT model from earlier GPT-6 Sol versions still used in Codex and ChatGPT Work. This chatbot configuration has no BMN quality verdict.

Next step & review dates

Original October7 announcement rechecked November17. Its Intelligent ChatGPT UI remains distinct from old API/Codex and November17 GPT6.2 Sol. Seek an exact UI task comparison; do not transfer newer release benchmarks.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

MicrosoftAwaiting testing access

MAI-Code-1.1-Flash

Announced Oct 7, 2026

Microsoft's October7 update describes downloadable MAI-Code-1.1-Flash with3-bit quantization and256K context, recommending more than120GB RAM; experimental on-device Copilot access is planned by month end. Performance, token-efficiency and price improvements are vendor claims, not a BMN quality verdict.

Next step & review dates

Original Microsoft access documentation rechecked November17. Code1.1Flash remains distinct from today's Code2/Code2Flash. Need the exact selectable client/build/effort and original matched repository comparison; new-release scores cannot be transferred.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Kandinsky LabGathering comparison evidence

Kandinsky 6.0

Announced Oct 6, 2026

Kandinsky Lab released the Kandinsky6.0 video/audio family on October6. Developer documentation describes Lite and Pro variants, five-second synchronized audio/video and optional Full-HD upscaling. Public weights and a fal Pro image-to-video API are documented; these are provider claims, not a BMN quality verdict.

Next step & review dates

In Voyager's Oct 22 native 480p fal trial, Pro preserved the supplied dripper but invented ice in the spill, distorted the title and changed the turntable product. Distinguish text generation from first-frame preservation. Next compare repeated takes on the same prompts at matched resolution; do not transfer results to Lite or the upscaler.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

RekaAwaiting public access

Rho-1

Announced Oct 5, 2026

Reka announced Rho-1, a 19B research preview unifying text, images, video and robotic actions in one network. Reka demonstrates continuous steerable video and describes the system as a proof of concept with limitations. Real-time capabilities are vendor claims; the base model is reported at 0.79× real-time, with faster distilled demonstrations. No BMN quality verdict.

Next step & review dates

Original request-access demonstration page rechecked November17; executable public exact route remains unverified. Confirm permitted usable configuration first; absence of access is not a quality verdict.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Fudan University / Tencent Hunyuan / Zhejiang University researchersGathering comparison evidence

Prism — research preview

Announced Oct 4, 2026

Prism research paper was disclosed October4. Original author-linked MIT preview_alpha (stable) and preview_beta (motion) weights and local CUDA inference instructions were verified October23. These are distinct research-preview configurations, not BMN-tested quality recommendations.

Next step & review dates

Original author GitHub README recovered despite inaccessible arXiv page. Preview alpha/beta checkpoints and CUDA inference support joint video/audio at documented resolutions; no quality test was run. Seek independent identical start-frame, seed, resolution and motion comparisons for each checkpoint.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Aleph AlphaAwaiting testing access

Kolibri

Announced Oct 3, 2026

Aleph Alpha released Kolibri, an Apache-2.0 open-weight English/German MoE with 78.1B total parameters, 3.46B active per token and support for up to a 1M-token context. Aleph Alpha reports 96.9 on AIME 2025 and 85.9 on LiveCodeBench v6; these are vendor-run claims, not a BMN quality verdict.

Next step & review dates

Original Apache2 weights and78B/3B-active metadata remain accessible. Hosted BMN writing route remains unverified. Seek original category-specific writing qualification and compliant exact serving route; any new screen also needs shared rolling admission/budget capacity. Vendor scores do not qualify.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

inclusionAIGathering comparison evidence

Ming Image 0.1 Design Layer

Announced Sep 23, 2026

Ant Ling’s Oct27 changelog verifies native API access. Layer takes a flattened image and returns decoupled PNG layers. OpenRouter lists a separate September catalog date; its live serving route was not verified.

Next step & review dates

Native access verified. Compare recovered alpha boundaries, foreground text and background reconstruction against a matched layer workflow.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

inclusionAIGathering comparison evidence

Ming Image 0.1 Design

Announced Sep 22, 2026

Ant Ling’s Oct27 changelog verifies native API access. Design supports prompt-to-image generation for graphic design. OpenRouter lists a separate September catalog date; its live serving route was not verified.

Next step & review dates

Native access verified. Seek original prompt-matched design comparisons for typography, layout and consistency; do not apply layer-edit results.

Original announcement ↗

Source checked Nov 17, 2026 · Next review target Nov 19, 2026

Past ranking and site updates
September 26, 2026

Opus 5.5 High enters the coding podium.

Today’s external-evidence review puts Claude Code + Opus 5.5 High first in Best Coding Overall and Code Generation, and second in Agentic Coding. These early recommendations carry an asterisk and a two-day review cycle. Exact effort settings, contrary results, and test limitations remain visible. Coding reviews proceed independently of pending writing approvals.

September 25, 2026

Clear recommendations. Reasons you can inspect.

Image and video reviews now support explicit first, second and third places, with category-specific criteria and published tiebreak decisions. Photorealism covers everyday phone snapshots as well as professional photography. Confidence and evidence gaps remain visible alongside our recommendations.

September 22, 2026

Image and video evidence, with its limits.

New source observations identify tested variants, task-specific findings, and conflicts with existing recommendations. They appear separately from the evidence behind the older rankings. Generic video summaries now receive the same provisional treatment as image summaries; scores and ranking-verification dates remain unchanged.

September 21, 2026

Clearer comparisons. Evidence you can inspect.

Larger text, direct category links, and expandable comparisons make the rankings easier to use. Evidence gaps are now visible, including inherited image placements awaiting version-specific confirmation. Ranking order and verification dates have not been advanced by this presentation update.

September 10, 2026

ChatGPT Images 2.5 replaces GPT Image 2.

OpenAI’s new image release adds faster generation, stronger subject preservation, more reliable multi-turn edits, and new direct-editing tools. The release initially inherited GPT Image 2’s placements. Those inherited placements are provisional until version-specific comparative evidence is documented. GPT-6 Astra’s access and API pricing have also been brought up to date.

September 8, 2026

Coding now ranks tested configurations.

Every Coding result identifies the system, underlying model or platform-selected AI, access plan, system version, and test date. Technical awards cleared the first supervised sweep; Builder awards are provisional until the full Practical Consensus threshold is met.

Evidence in.
A useful answer out.

Task-specific evidence, clear recommendations, and honest limits. Explore our criteria, sources, and how close calls are decided.

Read the full methodology →