Benchmark
Toolathlon Verified
Scores are percentages. Higher is better.
- Models and systems measured
- 48
- Scores recorded
- 75
- Sources
- 2
- IQ weight
- 0.90
Score distribution
- 59%
- Mean
- 55.6%
- Median
- 15.2%
- Std dev
- 39.8%
- 10th percentile
- 77.8%
- 90th percentile
- 25.6%
- Min
This benchmark has a weight of 0.90 for chat models, measured over 47 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model or system | Score |
|---|---|---|
| 1 | Claude Opus 5internal pinned harness · adaptive thinking · max · 108 tasks · 3 trials · Pass@1· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.6% |
| 2 | Claude Opus 4.8internal pinned harness · adaptive thinking · max · 108 tasks · 3 trials · Pass@1· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 79.9% |
| 3 | GLM 5.3 FlashVerified / 三次运行均值· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 78.4% |
| 4 | Claude Fable 5Qwen表:Verified · Pass@1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 77.9% |
| 5 | Claude Fable 5.1internal pinned harness · adaptive thinking · max · 108 tasks · 3 trials · Pass@1· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 77.8% |
| 6 | Claude Opus 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 77.8% |
| 7 | Claude Opus 5.5internal pinned harness · adaptive thinking · max · 108 tasks · 3 trials · Pass@1· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 77.8% |
| 8 | Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 77.8% |
| 9 | Claude Sonnet 5.5internal pinned harness · adaptive thinking · max · 108 tasks · 3 trials · Pass@1· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 77.8% |
| 10 | MiMo V2.6 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 76.9% |
| 11 | Kimi K3Verified · max effort · temp 1.0 · top_p 1.0· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76.5% |
| 12 | Kimi K3Verified / Pass@1· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76.5% |
| 13 | Claude Opus 4.8max · Toolathlon-Verified · official harness· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76.2% |
| 14 | Claude Opus 4.8Qwen表:Verified · Pass@1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76.2% |
| 15 | Claude Opus 4.8Verified / Pass@1· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76.2% |
| 16 | Claude Opus 4.8Verified / 三次运行均值· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76.2% |
| 17 | Muse Spark 1.1Meta Model API · xhigh · Toolathlon-Verified · official harness· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 75.6% |
| 18 | GPT-5.6 SolQwen表:Verified · Pass@1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.9% |
| 19 | GPT-5.6 SolVerified / Pass@1· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.9% |
| 20 | Claude Sonnet 5internal pinned harness · adaptive thinking · max · 108 tasks · 3 trials · Pass@1· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.7% |
| 21 | DeepSeek V4 Pro 0813Verified / Pass@1· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.1% |
| 22 | Hy4 previewHF Eval Result / Verified· Reported by Tencent Hy· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.1% |
| 23 | MiMo V2.6 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.6% |
| 24 | GPT 5.5xhigh · Toolathlon-Verified · official harness· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.5% |
| 25 | Qwen 3.8 Flash NextQwen3.8-Flash-Next release/model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.5% |
| 26 | GLM 5.3Verified / Pass@1· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73% |
| 27 | Muse Spark Open· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 72.8% |
| 28 | Qwen3.8 MaxQwen表:Verified · Pass@1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 72.5% |
| 29 | Qwen3.8 MaxVerified / Pass@1· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 72.5% |
| 30 | DeepSeek V4 Flash 0731Verified· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70.3% |
| 31 | Claude Sonnet 5internal pinned harness · pass@3 · 3 trials · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 63% |
| 32 | Claude Fable 5OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 61.7% |
| 33 | Gemini 3.1 Pro Previewhigh · Toolathlon-Verified · official harness· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 61.1% |
| 34 | Claude Opus 4.8OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 59.9% |
| 35 | GPT-5.6 SolOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 58% |
| 36 | Gemini 3.5 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.5% |
| 37 | GPT 5.5Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 55.6% |
| 38 | GPT 5.5OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 55.6% |
| 39 | GPT 5.5xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 55.6% |
| 40 | GPT 5.4MiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.6% |
| 41 | GPT 5.4OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.6% |
| 42 | Claude Sonnet 5internal pinned harness · pass@1 · 3 trials · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.3% |
| 43 | GPT-5.6 LunaOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.4% |
| 44 | GPT-5.6 TerraOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.1% |
| 45 | GPT-5.3 CodexOpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.9% |
| 46 | Kimi K2.6Kimi K2.6 official benchmark table · thinking · temp 1.0 · top_p 1.0 · 262,144 context· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 50% |
| 47 | Qwen3.7 MaxQwen表:Verified · Pass@1· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.7% |
| 48 | Step 3.7 FlashStep 3.7 Flash official README/figure· Reported by StepFun· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.5% |
| 49 | Claude Sonnet 4.6internal pinned harness · pass@1 · 3 trials · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.4% |
| 50 | Gemini 3 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.4% |
Showing 1–50 of 74 results.
Not ranked
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| Muse Spark (version unspecified)Meta Model API · highest effort · version unspecified · Toolathlon-Verified · official harness· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.4% |
Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats