Benchmark
SkillsBench · Vals · With Skills · Accuracy
Scores are percentages. Higher is better.
Benchmark sourceScores from BenchBoard docs.
- Models and systems measured
- 34
- Scores recorded
- 35
- Sources
- 1
- IQ weight
- 1.27
Score distribution
- 50.8%
- Mean
- 52.7%
- Median
- 11.6%
- Std dev
- 33.6%
- 10th percentile
- 62.6%
- 90th percentile
- 18.1%
- Min
This benchmark has a weight of 1.27 for chat models, measured over 24 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 | DeepSeek V4.1 FlashVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4.1-flash · provider=DeepSeek · reasoning_effort=high · max_output_tokens=384000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 69.8% |
| 2 | Grok 4.5Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=grok/grok-4.5 · provider=SpaceXAI · reasoning_effort=high · temperature=0.7 · top_p=0.95· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 66% |
| 3 | Gemini 3.7 FlashVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=google/gemini-3.7-flash · provider=Google · reasoning_effort=high · temperature=1 · max_output_tokens=65536· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 65.9% |
| 4 | GPT 5.5Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · Codex (version not stated) · With Skills · system=openai/gpt-5.5-codex · provider=OpenAI· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 62.6% |
| 5 | GPT 5.5Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.5 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 62.2% |
| 6 | Claude Fable 5.1Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-fable-5-1 · provider=Anthropic · compute_effort=max · temperature=1 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 61.6% |
| 7 | GPT-5.6 LunaVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.6-luna · provider=OpenAI · reasoning_effort=max · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 60.4% |
| 8 | Claude Opus 5Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-opus-5 · provider=Anthropic · compute_effort=max · temperature=1 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 60.4% |
| 9 | Claude Opus 4.8Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-opus-4-8 · provider=Anthropic · compute_effort=max · temperature=1 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 59.2% |
| 10 | Muse Spark 1.1Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=meta/muse_spark_1_1 · provider=Meta · reasoning_effort=xhigh · max_output_tokens=256000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 59.2% |
| 11 | GPT-5.6 TerraVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.6-terra · provider=OpenAI · reasoning_effort=max · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 58.9% |
| 12 | Gemini 3.8 FlashVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=google/gemini-3.8-flash · provider=Google · reasoning_effort=high · temperature=1 · max_output_tokens=65536· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 58% |
| 13 | Qwen3.7 PlusVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=alibaba/qwen3.7-plus · provider=Alibaba · temperature=0.7 · max_output_tokens=65536· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.3% |
| 14 | GPT-5.6 SolVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.6-sol · provider=OpenAI · reasoning_effort=max · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.1% |
| 15 | DeepSeek V4 Pro 0813Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4-pro-0813 · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.8% |
| 16 | Muse Spark 1.2Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53% |
| 17 | Gemini 3.5 FlashVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=google/gemini-3.5-flash · provider=Google · reasoning_effort=high · temperature=1 · max_output_tokens=65536· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 52.7% |
| 18 | GPT 5.4Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.7% |
| 19 | MiniMax M3Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=minimax/MiniMax-M3 · provider=MiniMax · temperature=1 · top_p=0.95 · max_output_tokens=512000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.5% |
| 20 | DeepSeek V4 Flash 0731Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4-flash-0731 · provider=DeepSeek · reasoning_effort=high · max_output_tokens=384000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 50.7% |
| 21 | Kimi K2.7 CodeVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=kimi/kimi-k2.7-code · provider=Moonshot AI · top_p=0.95 · max_output_tokens=32768· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 50% |
| 22 | Claude Sonnet 4.6Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-sonnet-4-6 · provider=Anthropic · compute_effort=max · reasoning=true · temperature=1 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.1% |
| 23 | GLM 5.3Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=zai/glm-5.3 · provider=Zhipu AI · reasoning_effort=max · temperature=1 · top_p=0.95 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 47.5% |
| 24 | Claude Sonnet 5Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=anthropic/claude-sonnet-5 · provider=Anthropic · compute_effort=max · temperature=1 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 46.5% |
| 25 | GLM 5.2Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=zai/glm-5.2 · provider=Zhipu AI · reasoning_effort=max · temperature=1 · top_p=0.95 · max_output_tokens=131072· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 45.1% |
| 26 | Qwen3.8 MaxVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=alibaba/qwen3.8-max · provider=Alibaba · temperature=0.7 · max_output_tokens=128000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 42% |
| 27 | GLM 5.3 FlashVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=zai/glm-5.3-flash · provider=Fireworks AI · reasoning_effort=max · temperature=1 · top_p=0.95 · max_output_tokens=131072· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 40.2% |
| 28 | Qwen 3.8 27BVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=alibaba/qwen3.8-27b · provider=Alibaba · reasoning_effort=xhigh · temperature=1 · top_p=0.95 · max_output_tokens=65536· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 38.1% |
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 |
|---|---|
| DeepSeek V4 Pro (version unspecified)Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=deepseek/deepseek-v4-pro · provider=DeepSeek · reasoning_effort=max · max_output_tokens=384000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.3% |
| Grok 4.3Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=grok/grok-4.3 · provider=SpaceXAI · reasoning_effort=high · temperature=0.7 · top_p=0.95· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 40.6% |
| Grok 4.6Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=grok/grok-4.6 · provider=SpaceXAI · reasoning_effort=high · temperature=0.7 · top_p=0.95· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 55.8% |
| InklingVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=thinkingmachines/inkling · provider=Thinkingmachines · reasoning_effort=0.99 · temperature=1 · top_p=1 · max_output_tokens=256000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 26.1% |
| Inkling-SmallVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=thinkingmachines/inkling-small · provider=Thinkingmachines · reasoning_effort=0.99 · temperature=1 · top_p=1 · max_output_tokens=262000· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 33.6% |
| Ling-3.0-FlashVals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=ant/ling-3.0-flash-2607 · provider=Ant · temperature=0.6 · top_p=0.95 · max_output_tokens=131072· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 27.9% |
| Mercury 2.5Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=inception/mercury-2.5 · provider=Inception · reasoning_effort=high · max_output_tokens=65536· Reported by Vals AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 18.1% |