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. Claude Fable 5.176.78Independent

    Anthropic

  2. GPT-6 Astra75.43Independent

    OpenAI

  3. Gemini 4 Argon72.21Independent

    Google DeepMind

  4. GPT-6.1 Sol70.89Independent

    OpenAI

  5. Claude Opus 570.66Independent

    Anthropic

  6. Mistral Large 4Not reported

    Mistral AI

Agentic

BenchLM · 0 to 100

  1. Claude Fable 5.180.94Independent

    Anthropic

  2. Claude Opus 578.55Independent

    Anthropic

  3. Gemini 4 Argon77.28Independent

    Google DeepMind

  4. GPT-6.1 Sol76.27Independent

    OpenAI

  5. GPT-6 Astra74.93Independent

    OpenAI

  6. Mistral Large 454.07Independent

    Mistral AI

Reasoning

BenchLM · 0 to 100

  1. GPT-6 Astra93.3Independent

    OpenAI

  2. GPT-6.1 Sol90.5Independent

    OpenAI

  3. Claude Fable 5.185.2Independent

    Anthropic

  4. Mistral Large 481.7Independent

    Mistral AI

  5. Claude Opus 581Independent

    Anthropic

  6. Gemini 4 ArgonNot reported

    Google DeepMind

Knowledge

BenchLM · 0 to 100

  1. Claude Fable 5.186.37Independent

    Anthropic

  2. GPT-6 Astra85.69Independent

    OpenAI

  3. GPT-6.1 Sol85.56Independent

    OpenAI

  4. Claude Opus 581.59Independent

    Anthropic

  5. Gemini 4 Argon80.01Independent

    Google DeepMind

  6. Mistral Large 4Not reported

    Mistral AI

Arena

LM Arena · Elo

  1. Google DeepMind

  2. Anthropic

  3. Claude Opus 51,489.85Independent

    Anthropic

  4. GPT-6.1 Sol1,483.96Independent

    OpenAI

  5. GPT-6 Astra1,476.72Independent

    OpenAI

  6. Mistral AI

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 Fable 5.1, Coding: 76.78. Independent. Captured 2026-10-15.Claude Fable 5.1, Agentic: 80.94. Independent. Captured 2026-10-15.Claude Fable 5.1, Reasoning: 85.2. Independent. Captured 2026-10-15.Claude Fable 5.1, Knowledge: 86.37. Independent. Captured 2026-10-15.Claude Fable 5.1, Arena: 1,500.57. Independent. Captured 2026-10-15.GPT-6 Astra, Coding: 75.43. Independent. Captured 2026-10-15.GPT-6 Astra, Agentic: 74.93. Independent. Captured 2026-10-15.GPT-6 Astra, Reasoning: 93.3. Independent. Captured 2026-10-15.GPT-6 Astra, Knowledge: 85.69. Independent. Captured 2026-10-15.GPT-6 Astra, Arena: 1,476.72. Independent. Captured 2026-10-15.GPT-6.1 Sol, Coding: 70.89. Independent. Captured 2026-10-15.GPT-6.1 Sol, Agentic: 76.27. Independent. Captured 2026-10-15.GPT-6.1 Sol, Reasoning: 90.5. Independent. Captured 2026-10-15.GPT-6.1 Sol, Knowledge: 85.56. Independent. Captured 2026-10-15.GPT-6.1 Sol, Arena: 1,483.96. Independent. Captured 2026-10-15.

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 Argon

Released 2026-10-14 · 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
+5.54pts
71.24 to 76.78 Independent2026-10-15 / 2026-10-15
Agentic
+4.97pts
75.97 to 80.94 Independent2026-10-15 / 2026-10-15
Knowledge
+4.28pts
82.09 to 86.37 Independent2026-10-15 / 2026-10-15
Arena
-2.81Elo
1,503.38 to 1,500.57 Independent2026-10-15 / 2026-10-15

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
Claude Fable 5.1CodingBenchLM.ai76.78Independent2026-10-15Claude Fable 5.1
GPT-6 AstraCodingBenchLM.ai75.43Independent2026-10-15GPT-6 Astra
Gemini 4 ArgonCodingBenchLM.ai72.21Independent2026-10-15Gemini 4 Argon
GPT-6.1 SolCodingBenchLM.ai70.89Independent2026-10-15GPT-6.1 Sol
Claude Opus 5CodingBenchLM.ai70.66Independent2026-10-15Claude Opus 5
Mistral Large 4CodingBenchLM.aiNot reportedNot reportedNot reported
Claude Fable 5.1AgenticBenchLM.ai80.94Independent2026-10-15Claude Fable 5.1
Claude Opus 5AgenticBenchLM.ai78.55Independent2026-10-15Claude Opus 5
Gemini 4 ArgonAgenticBenchLM.ai77.28Independent2026-10-15Gemini 4 Argon
GPT-6.1 SolAgenticBenchLM.ai76.27Independent2026-10-15GPT-6.1 Sol
GPT-6 AstraAgenticBenchLM.ai74.93Independent2026-10-15GPT-6 Astra
Mistral Large 4AgenticBenchLM.ai54.07Independent2026-10-14Mistral Large 4
GPT-6 AstraReasoningBenchLM.ai93.3Independent2026-10-15GPT-6 Astra
GPT-6.1 SolReasoningBenchLM.ai90.5Independent2026-10-15GPT-6.1 Sol
Claude Fable 5.1ReasoningBenchLM.ai85.2Independent2026-10-15Claude Fable 5.1
Mistral Large 4ReasoningBenchLM.ai81.7Independent2026-10-14Mistral Large 4
Claude Opus 5ReasoningBenchLM.ai81Independent2026-10-15Claude Opus 5
Gemini 4 ArgonReasoningBenchLM.aiNot reportedNot reportedNot reported
Claude Fable 5.1KnowledgeBenchLM.ai86.37Independent2026-10-15Claude Fable 5.1
GPT-6 AstraKnowledgeBenchLM.ai85.69Independent2026-10-15GPT-6 Astra
GPT-6.1 SolKnowledgeBenchLM.ai85.56Independent2026-10-15GPT-6.1 Sol
Claude Opus 5KnowledgeBenchLM.ai81.59Independent2026-10-15Claude Opus 5
Gemini 4 ArgonKnowledgeBenchLM.ai80.01Independent2026-10-15Gemini 4 Argon
Mistral Large 4KnowledgeBenchLM.aiNot reportedNot reportedNot reported
Gemini 4 ArgonArenaLM Arena1,524.77Independent2026-10-15gemini-4-argon-high4,903 votes
Claude Fable 5.1ArenaLM Arena1,500.57Independent2026-10-15claude-fable-5.1-max14,121 votes
Claude Opus 5ArenaLM Arena1,489.85Independent2026-10-15claude-opus-5-high65,663 votes
GPT-6.1 SolArenaLM Arena1,483.96Independent2026-10-15gpt-6.1-sol-max5,762 votes
GPT-6 AstraArenaLM Arena1,476.72Independent2026-10-15gpt-6-astra-max11,674 votes
Mistral Large 4ArenaLM Arena1,431.48Independent2026-10-15mistral-large-46,607 votes
Additional readings for these models. Separate benchmarks, not substitutes for the five chart axes.
ModelMetric / sourceScore / provenanceCaptured
Claude Fable 5.1Humanity's Last ExamEpoch AI46.5Independent2026-10-15
Claude Fable 5.1GDP.pdfEpoch AI29.6Independent2026-10-15
GPT-6 AstraHumanity's Last ExamEpoch AI54.8Independent2026-10-15
Claude Opus 5MultimodalGroundedBenchLM.ai92.2Independent2026-10-15
Claude Opus 5GDP.pdfEpoch AI24Independent2026-10-15

Mirrored snapshots, not live scores. Epoch AI values retain their original scale; chart geometry is normalized. Evaluation dates, source labels and sample sizes may differ.

Open page →
Provider
Meta
-
Out yet?
Yes
-
Status
Available
-
Release date
Aug 9, 2026
-
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
-
-