Benchmarks

Compare AI models

Pick two models to see release dates, benchmark scores, and community ratings side by side.

Frontier benchmarks

Frontier board

6 releases. Five ways to compare.

Selected for recent releases and benchmark coverage, not a single overall rank. Each column sorts the same models by its own score. No reading means no bar.

○ Independent evaluation · ◆ Vendor or self-reported, labeled per reading.

Coding

BenchLM · 0 to 100

  1. Gemini 4 Carbon83.97Independent

    Google DeepMind

  2. Claude Opus 5.582.67Independent

    Anthropic

  3. GPT-6 Astra78.73Independent

    OpenAI

  4. Claude Fable 5.173.98Independent

    Anthropic

  5. GPT-6.1 Sol73.39Independent

    OpenAI

  6. Gemini 4 Argon71.88Independent

    Google DeepMind

Agentic

BenchLM · 0 to 100

  1. Gemini 4 Carbon84.64Independent

    Google DeepMind

  2. Claude Opus 5.583.85Independent

    Anthropic

  3. Claude Fable 5.179.21Independent

    Anthropic

  4. Gemini 4 Argon74.86Independent

    Google DeepMind

  5. GPT-6 Astra73.65Independent

    OpenAI

  6. GPT-6.1 Sol65.66Independent

    OpenAI

Reasoning

BenchLM · 0 to 100

  1. Gemini 4 Carbon84.84Independent

    Google DeepMind

  2. GPT-6 Astra81.12Independent

    OpenAI

  3. Claude Opus 5.580.53Independent

    Anthropic

  4. GPT-6.1 Sol80.41Independent

    OpenAI

  5. Gemini 4 Argon76.52Independent

    Google DeepMind

  6. Claude Fable 5.176.36Independent

    Anthropic

Knowledge

BenchLM · 0 to 100

  1. Claude Opus 5.589.75Independent

    Anthropic

  2. GPT-6.1 Sol85.57Independent

    OpenAI

  3. GPT-6 Astra85.45Independent

    OpenAI

  4. Claude Fable 5.184.08Independent

    Anthropic

  5. Gemini 4 Argon79.21Independent

    Google DeepMind

  6. Gemini 4 CarbonNot reported

    Google DeepMind

Arena

LM Arena · Elo

  1. Google DeepMind

  2. Anthropic

  3. Anthropic

  4. GPT-6.1 Sol1,479.94Independent

    OpenAI

  5. GPT-6 Astra1,476.48Independent

    OpenAI

  6. Gemini 4 CarbonNot reported

    Google DeepMind

Score links open the source; hover for capture details. Bar lengths start at zero. Arena lengths use the full public board maximum.

The shape of the scores

First three featured models. Toggle a name to isolate a line. Missing readings leave gaps, not zeroes.

CodingAgenticReasoningKnowledgeArena050100Claude Opus 5.5, Coding: 82.67. Independent. Captured 2026-11-13.Claude Opus 5.5, Agentic: 83.85. Independent. Captured 2026-11-13.Claude Opus 5.5, Reasoning: 80.53. Independent. Captured 2026-11-13.Claude Opus 5.5, Knowledge: 89.75. Independent. Captured 2026-11-13.Claude Opus 5.5, Arena: 1,508.73. Independent. Captured 2026-10-26.GPT-6 Astra, Coding: 78.73. Independent. Captured 2026-11-13.GPT-6 Astra, Agentic: 73.65. Independent. Captured 2026-11-13.GPT-6 Astra, Reasoning: 81.12. Independent. Captured 2026-11-13.GPT-6 Astra, Knowledge: 85.45. Independent. Captured 2026-11-13.GPT-6 Astra, Arena: 1,476.48. Independent. Captured 2026-10-26.Gemini 4 Carbon, Coding: 83.97. Independent. Captured 2026-11-13.Gemini 4 Carbon, Agentic: 84.64. Independent. Captured 2026-11-13.Gemini 4 Carbon, Reasoning: 84.84. Independent. Captured 2026-11-13.

3 of 3 series shown.

Axes share drawing space, not meaning. Arena is Elo divided by the current board maximum; it is not a percentage accuracy.

The newest arrival

Gemini 4 Carbon

Released 2026-11-10 · At least two readings

No predecessor is linked in the catalog yet. We do not infer a generation jump from a name.

Latest release with linked lineage

Claude Fable 5.1 vs Claude Fable 5

Coding
+6.16pts
67.82 to 73.98 Independent2026-11-13 / 2026-11-13
Agentic
+6.54pts
72.67 to 79.21 Independent2026-11-13 / 2026-11-13
Reasoning
+4.72pts
71.64 to 76.36 Independent2026-11-13 / 2026-11-13
Knowledge
+4.64pts
79.44 to 84.08 Independent2026-11-13 / 2026-11-13
Arena
-0.09Elo
1,503.28 to 1,503.19 Independent2026-10-26 / 2026-10-26

Latest snapshots, not a controlled before-and-after trial. Capture dates and evaluation setups may differ.

Scores and sources

Frontier board: original scores. Not reported is missing, not zero.
ModelMetricScore / provenanceCapturedSource label / votes
Gemini 4 CarbonCodingBenchLM.ai83.97Independent2026-11-13Gemini 4 Carbon
Claude Opus 5.5CodingBenchLM.ai82.67Independent2026-11-13Claude Opus 5.5
GPT-6 AstraCodingBenchLM.ai78.73Independent2026-11-13GPT-6 Astra
Claude Fable 5.1CodingBenchLM.ai73.98Independent2026-11-13Claude Fable 5.1
GPT-6.1 SolCodingBenchLM.ai73.39Independent2026-11-13GPT-6.1 Sol
Gemini 4 ArgonCodingBenchLM.ai71.88Independent2026-11-13Gemini 4 Argon
Gemini 4 CarbonAgenticBenchLM.ai84.64Independent2026-11-13Gemini 4 Carbon
Claude Opus 5.5AgenticBenchLM.ai83.85Independent2026-11-13Claude Opus 5.5
Claude Fable 5.1AgenticBenchLM.ai79.21Independent2026-11-13Claude Fable 5.1
Gemini 4 ArgonAgenticBenchLM.ai74.86Independent2026-11-13Gemini 4 Argon
GPT-6 AstraAgenticBenchLM.ai73.65Independent2026-11-13GPT-6 Astra
GPT-6.1 SolAgenticBenchLM.ai65.66Independent2026-11-13GPT-6.1 Sol
Gemini 4 CarbonReasoningBenchLM.ai84.84Independent2026-11-13Gemini 4 Carbon
GPT-6 AstraReasoningBenchLM.ai81.12Independent2026-11-13GPT-6 Astra
Claude Opus 5.5ReasoningBenchLM.ai80.53Independent2026-11-13Claude Opus 5.5
GPT-6.1 SolReasoningBenchLM.ai80.41Independent2026-11-13GPT-6.1 Sol
Gemini 4 ArgonReasoningBenchLM.ai76.52Independent2026-11-13Gemini 4 Argon
Claude Fable 5.1ReasoningBenchLM.ai76.36Independent2026-11-13Claude Fable 5.1
Claude Opus 5.5KnowledgeBenchLM.ai89.75Independent2026-11-13Claude Opus 5.5
GPT-6.1 SolKnowledgeBenchLM.ai85.57Independent2026-11-13GPT-6.1 Sol
GPT-6 AstraKnowledgeBenchLM.ai85.45Independent2026-11-13GPT-6 Astra
Claude Fable 5.1KnowledgeBenchLM.ai84.08Independent2026-11-13Claude Fable 5.1
Gemini 4 ArgonKnowledgeBenchLM.ai79.21Independent2026-11-13Gemini 4 Argon
Gemini 4 CarbonKnowledgeBenchLM.aiNot reportedNot reportedNot reported
Gemini 4 ArgonArenaLM Arena1,517.17Independent2026-10-26Gemini 4 Argon (High)10,083 votes
Claude Opus 5.5ArenaLM Arena1,508.73Independent2026-10-26Claude Opus 5.5 (High)10,322 votes
Claude Fable 5.1ArenaLM Arena1,503.19Independent2026-10-26Claude Fable 5.1 (Max)16,343 votes
GPT-6.1 SolArenaLM Arena1,479.94Independent2026-10-26GPT-6.1 Sol (Max)8,305 votes
GPT-6 AstraArenaLM Arena1,476.48Independent2026-10-26GPT-6 Astra (Max)14,076 votes
Gemini 4 CarbonArenaLM ArenaNot reportedNot reportedNot reported
Additional readings for these models. Separate benchmarks, not substitutes for the five chart axes.
ModelMetric / sourceScore / provenanceCaptured
Claude Opus 5.5MultimodalGroundedBenchLM.ai85.09Independent2026-11-13
Claude Opus 5.5GDP.pdfEpoch AI30.6Independent2026-11-13
GPT-6 AstraHumanity's Last ExamEpoch AI54.8Independent2026-11-13
GPT-6 AstraMultimodalGroundedBenchLM.ai79.7Independent2026-11-13
Claude Fable 5.1Humanity's Last ExamEpoch AI46.5Independent2026-11-13
Claude Fable 5.1GDP.pdfEpoch AI29.6Independent2026-11-13
Claude Fable 5.1MultimodalGroundedBenchLM.ai80.54Independent2026-11-13
Gemini 4 ArgonGDP.pdfEpoch AI28.4Independent2026-11-13
GPT-6.1 SolMultimodalGroundedBenchLM.ai78.88Independent2026-11-13
Open page →
Provider
ByteDance Seed
-
Out yet?
No
-
Status
Speculated
-
Release date
-
-
Vibe rating?OutYet's community rating: signed-in users score the model 1–10. Shown as the average and the number of votes.
-
-
Predecessor
-
-
Successor
-
-