OutYet.ai

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 Opus 5.582.62Independent

    Anthropic

  2. Claude Fable 5.176.81Independent

    Anthropic

  3. GPT-6 Astra75.46Independent

    OpenAI

  4. Gemini 4 Argon72.28Independent

    Google DeepMind

  5. GPT-6.1 Sol70.93Independent

    OpenAI

  6. Muse Spark Open56.08Independent

    Meta

Agentic

BenchLM · 0 to 100

  1. Claude Opus 5.590.31Independent

    Anthropic

  2. Claude Fable 5.180.86Independent

    Anthropic

  3. Gemini 4 Argon77.23Independent

    Google DeepMind

  4. GPT-6.1 Sol76.18Independent

    OpenAI

  5. GPT-6 Astra74.8Independent

    OpenAI

  6. Muse Spark Open60.08Independent

    Meta

Reasoning

BenchLM · 0 to 100

  1. GPT-6 Astra93.3Independent

    OpenAI

  2. GPT-6.1 Sol90.5Independent

    OpenAI

  3. Claude Opus 5.586.1Independent

    Anthropic

  4. Claude Fable 5.185.2Independent

    Anthropic

  5. Gemini 4 ArgonNot reported

    Google DeepMind

  6. Muse Spark OpenNot reported

    Meta

Knowledge

BenchLM · 0 to 100

  1. Claude Opus 5.590.06Independent

    Anthropic

  2. Claude Fable 5.186.75Independent

    Anthropic

  3. GPT-6 Astra86.02Independent

    OpenAI

  4. GPT-6.1 Sol85.9Independent

    OpenAI

  5. Gemini 4 Argon80.31Independent

    Google DeepMind

  6. Muse Spark Open64.6Independent

    Meta

Arena

LM Arena · Elo

  1. Gemini 4 Argon1,523.9Independent

    Google DeepMind

  2. Claude Opus 5.51,508.17Independent

    Anthropic

  3. Claude Fable 5.11,502.29Independent

    Anthropic

  4. GPT-6.1 Sol1,483.85Independent

    OpenAI

  5. GPT-6 Astra1,477.32Independent

    OpenAI

  6. Muse Spark OpenNot reported

    Meta

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: 82.62. Independent. Captured 2026-10-17.Claude Opus 5.5, Agentic: 90.31. Independent. Captured 2026-10-17.Claude Opus 5.5, Reasoning: 86.1. Independent. Captured 2026-10-17.Claude Opus 5.5, Knowledge: 90.06. Independent. Captured 2026-10-17.Claude Opus 5.5, Arena: 1,508.17. Independent. Captured 2026-10-17.Claude Fable 5.1, Coding: 76.81. Independent. Captured 2026-10-17.Claude Fable 5.1, Agentic: 80.86. Independent. Captured 2026-10-17.Claude Fable 5.1, Reasoning: 85.2. Independent. Captured 2026-10-17.Claude Fable 5.1, Knowledge: 86.75. Independent. Captured 2026-10-17.Claude Fable 5.1, Arena: 1,502.29. Independent. Captured 2026-10-17.GPT-6 Astra, Coding: 75.46. Independent. Captured 2026-10-17.GPT-6 Astra, Agentic: 74.8. Independent. Captured 2026-10-17.GPT-6 Astra, Reasoning: 93.3. Independent. Captured 2026-10-17.GPT-6 Astra, Knowledge: 86.02. Independent. Captured 2026-10-17.GPT-6 Astra, Arena: 1,477.32. Independent. Captured 2026-10-17.

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

Muse Spark Open

Released 2026-10-15 · 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
+5.55pts
71.26 to 76.81 Independent2026-10-17 / 2026-10-17
Agentic
+4.98pts
75.88 to 80.86 Independent2026-10-17 / 2026-10-17
Knowledge
+4.35pts
82.4 to 86.75 Independent2026-10-17 / 2026-10-17
Arena
-1.32Elo
1,503.6 to 1,502.29 Independent2026-10-17 / 2026-10-17

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 Opus 5.5CodingBenchLM.ai82.62Independent2026-10-17Claude Opus 5.5
Claude Fable 5.1CodingBenchLM.ai76.81Independent2026-10-17Claude Fable 5.1
GPT-6 AstraCodingBenchLM.ai75.46Independent2026-10-17GPT-6 Astra
Gemini 4 ArgonCodingBenchLM.ai72.28Independent2026-10-17Gemini 4 Argon
GPT-6.1 SolCodingBenchLM.ai70.93Independent2026-10-17GPT-6.1 Sol
Muse Spark OpenCodingBenchLM.ai56.08Independent2026-10-17Muse Spark Open
Claude Opus 5.5AgenticBenchLM.ai90.31Independent2026-10-17Claude Opus 5.5
Claude Fable 5.1AgenticBenchLM.ai80.86Independent2026-10-17Claude Fable 5.1
Gemini 4 ArgonAgenticBenchLM.ai77.23Independent2026-10-17Gemini 4 Argon
GPT-6.1 SolAgenticBenchLM.ai76.18Independent2026-10-17GPT-6.1 Sol
GPT-6 AstraAgenticBenchLM.ai74.8Independent2026-10-17GPT-6 Astra
Muse Spark OpenAgenticBenchLM.ai60.08Independent2026-10-17Muse Spark Open
GPT-6 AstraReasoningBenchLM.ai93.3Independent2026-10-17GPT-6 Astra
GPT-6.1 SolReasoningBenchLM.ai90.5Independent2026-10-17GPT-6.1 Sol
Claude Opus 5.5ReasoningBenchLM.ai86.1Independent2026-10-17Claude Opus 5.5
Claude Fable 5.1ReasoningBenchLM.ai85.2Independent2026-10-17Claude Fable 5.1
Gemini 4 ArgonReasoningBenchLM.aiNot reportedNot reportedNot reported
Muse Spark OpenReasoningBenchLM.aiNot reportedNot reportedNot reported
Claude Opus 5.5KnowledgeBenchLM.ai90.06Independent2026-10-17Claude Opus 5.5
Claude Fable 5.1KnowledgeBenchLM.ai86.75Independent2026-10-17Claude Fable 5.1
GPT-6 AstraKnowledgeBenchLM.ai86.02Independent2026-10-17GPT-6 Astra
GPT-6.1 SolKnowledgeBenchLM.ai85.9Independent2026-10-17GPT-6.1 Sol
Gemini 4 ArgonKnowledgeBenchLM.ai80.31Independent2026-10-17Gemini 4 Argon
Muse Spark OpenKnowledgeBenchLM.ai64.6Independent2026-10-17Muse Spark Open
Gemini 4 ArgonArenaLM Arena1,523.9Independent2026-10-17Gemini 4 Argon (High)4,904 votes
Claude Opus 5.5ArenaLM Arena1,508.17Independent2026-10-17Claude Opus 5.5 (High)8,300 votes
Claude Fable 5.1ArenaLM Arena1,502.29Independent2026-10-17Claude Fable 5.1 (Max)14,590 votes
GPT-6.1 SolArenaLM Arena1,483.85Independent2026-10-17GPT-6.1 Sol (Max)6,320 votes
GPT-6 AstraArenaLM Arena1,477.32Independent2026-10-17GPT-6 Astra (Max)12,163 votes
Muse Spark OpenArenaLM 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.ai92.2Independent2026-10-17
Claude Opus 5.5GDP.pdfEpoch AI30.6Independent2026-10-17
Claude Fable 5.1Humanity's Last ExamEpoch AI46.5Independent2026-10-17
Claude Fable 5.1GDP.pdfEpoch AI29.6Independent2026-10-17
GPT-6 AstraHumanity's Last ExamEpoch AI54.8Independent2026-10-17
Open page →
Provider
Anthropic
-
Out yet?
Yes
-
Status
Available
-
Release date
Sep 22, 2026
-
LM Arena?LM Arena's Elo-style rating from blind head-to-head votes: people compare two anonymous model answers and pick the better one. Higher is better.
1508.2Verify at LM ArenaOct 17, 2026
-
Agentic?BenchLM.ai's score for multi-step agentic work - planning, tool use, and acting autonomously, normalized 0–100 across multiple benchmarks. Higher is better.
90.31Verify at BenchLM.aiOct 17, 2026
-
Coding?BenchLM.ai's score for code generation and software-engineering tasks, normalized 0–100 across multiple benchmarks. Higher is better.
82.62Verify at BenchLM.aiOct 17, 2026
-
Knowledge?BenchLM.ai's score for factual knowledge and question answering, normalized 0–100 across multiple benchmarks. Higher is better.
90.06Verify at BenchLM.aiOct 17, 2026
-
MultimodalGrounded?BenchLM.ai's score for understanding grounded in images and documents, normalized 0–100 across multiple benchmarks. Higher is better.
92.2Verify at BenchLM.aiOct 17, 2026
-
Reasoning?BenchLM.ai's score for logic and multi-step reasoning problems, normalized 0–100 across multiple benchmarks. Higher is better.
86.1Verify at BenchLM.aiOct 17, 2026
-
GDP.pdf?Epoch AI's GDP.pdf benchmark: extracting and analyzing information from real-world PDF documents. Higher is better.Epoch AI
30.6Verify at Epoch AIOct 17, 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
-
-

Benchmark scores are mirrored from third-party sources and captured on the dates shown. Numbers from different benchmarks, sources, or capture dates are not directly comparable.

Data from BenchLM.ai · Epoch AI, “AI Benchmarking Hub” (CC BY 4.0).