benchgap
Agentic · tools

QwenClawBench leaderboard

As of 2026-11-22, the highest measured score on QwenClawBench is 64.3% by Qwen3.7 Max. 127 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#ModelScoreSource
1Muse Spark 1.293.2%estimated ± 6.0 pp, low confidence
2Claude Sonnet 593.2%estimated ± 6.0 pp, low confidence
3GPT-5.6 Luna93.2%estimated ± 6.0 pp, low confidence
4MiniMax M3.1 Flash93.2%estimated ± 6.0 pp, low confidence
5GPT-5.6 Terra93.2%estimated ± 6.0 pp, low confidence
6Gemini 3.8 Flash93.2%estimated ± 6.0 pp, low confidence
7GPT-6 Luna93.2%estimated ± 6.0 pp, low confidence
8Mistral Large 493.2%estimated ± 6.0 pp, low confidence
9Qwen3.8-27B93.2%estimated ± 6.0 pp, low confidence
10Gemini 3.7 Flash93.2%estimated ± 6.0 pp, low confidence
11Grok 4.593.2%estimated ± 6.0 pp, low confidence
12Gemini 3.6 Flash93.2%estimated ± 6.0 pp, low confidence
13Hy3 Preview93.1%estimated ± 6.0 pp, low confidence
14Apodex 1.192.2%estimated ± 6.0 pp, low confidence
15Apodex 1.1 Mini92.2%estimated ± 6.0 pp, low confidence
16Quasar 438B91.6%estimated ± 6.0 pp, low confidence
17Nemotron 3.5 Super VL 120B A12B86.4%estimated ± 6.0 pp, low confidence
18Nemotron 3.5 Super VL 120B A12B NVFP486.4%estimated ± 6.0 pp, low confidence
19Nemotron 3 Ultra86.4%estimated ± 6.0 pp, low confidence
20Ling 3.0 Flash VL85.4%estimated ± 6.0 pp, low confidence
21Muse Spark Open74.1%estimated ± 2.6 pp, medium confidence
22Muse Spark 1.173.0%estimated ± 2.6 pp, medium confidence
23Claude Opus 570.9%estimated ± 2.6 pp, medium confidence
24Step 5 Preview70.7%estimated ± 2.6 pp, medium confidence
25Inkling 2 Small70.2%estimated ± 2.6 pp, medium confidence
26ERNIE 5.5 Flash Preview69.8%estimated ± 2.6 pp, medium confidence
27Kimi K369.4%estimated ± 2.6 pp, medium confidence
28Hy4 preview69.0%estimated ± 2.6 pp, medium confidence
29Gemini 3.5 Flash68.9%estimated ± 2.6 pp, medium confidence
30LongCat-2.568.5%estimated ± 2.6 pp, medium confidence
31Ling 3.1 Flash68.4%estimated ± 2.6 pp, medium confidence
32Inkling 2 Mini68.1%estimated ± 2.6 pp, medium confidence
33Claude Opus 4.867.7%estimated ± 2.6 pp, medium confidence
34Beam 1.0 501B-A23B FP867.1%estimated ± 2.6 pp, medium confidence
35Beam 1.0 501B-A23B66.6%estimated ± 2.6 pp, medium confidence
36Beam 1.0 501B-A23B NVFP466.3%estimated ± 2.6 pp, medium confidence
37Ornith-1.5-397B65.9%estimated ± 2.6 pp, medium confidence
38Inkling-Small65.6%estimated ± 2.6 pp, medium confidence
39Qwen3.7 Max64.3%measured
40Claude Opus 4.763.9%estimated ± 2.6 pp, medium confidence
41GLM-5.263.6%estimated ± 2.6 pp, medium confidence
42Kimi K2.7 Code63.0%estimated ± 2.6 pp, high confidence
43Ling 3.0 Flash Sante62.8%estimated ± 2.6 pp, high confidence
44Muse Glimmer 30B62.7%estimated ± 2.6 pp, high confidence
45GPT-5.562.6%estimated ± 2.6 pp, high confidence
46MiniMax M361.9%estimated ± 2.6 pp, high confidence
47Inkling61.8%estimated ± 2.6 pp, high confidence
48Qwen3.7 Plus61.8%measured
49Sage-A1.561.8%estimated ± 2.0 pp, medium confidence
50DeepSeek V4 Pro61.5%estimated ± 2.6 pp, high confidence
51MiMo-V2.5-Pro60.8%estimated ± 6.0 pp, low confidence
52GLM-5.160.5%estimated ± 2.6 pp, high confidence
53Ripple-65B Preview60.4%estimated ± 2.6 pp, high confidence
54GPT-5.459.8%estimated ± 2.6 pp, high confidence
55Ornith-1.5-35B-A3B59.6%estimated ± 2.6 pp, high confidence
56DeepSeek V4 Flash 073159.1%estimated ± 2.6 pp, high confidence
57Qwen 3.6 Max (preview)59.0%measured
58Ling 3.0 Flash57.6%estimated ± 2.6 pp, high confidence
59Qwen3.6 Plus57.2%measured
60Spark-X2.6-4B56.5%estimated ± 2.6 pp, high confidence
61Solar Pro 456.2%estimated ± 2.6 pp, high confidence
62Grok 4.356.2%estimated ± 6.0 pp, low confidence
63Hy355.9%estimated ± 6.0 pp, low confidence
64Step 3.7 Flash55.9%estimated ± 6.0 pp, low confidence
65MiniMax M2.755.9%estimated ± 6.0 pp, low confidence
66Muse Spark55.9%estimated ± 6.0 pp, low confidence
67Gemini 3.5 Flash-Lite55.9%estimated ± 6.0 pp, low confidence
68A.X K255.9%estimated ± 6.0 pp, low confidence
69GPT-555.9%estimated ± 6.0 pp, low confidence
70GPT-5.155.9%estimated ± 6.0 pp, low confidence
71Qwen3.5-122B-A10B55.9%estimated ± 6.0 pp, low confidence
72K-EXAONE 2.055.9%estimated ± 6.0 pp, low confidence
73Gemini 3.1 Pro55.9%estimated ± 6.0 pp, low confidence
74Mistral Medium 3.5 128B55.9%estimated ± 6.0 pp, low confidence
75MiniCPM5-2B55.9%estimated ± 6.0 pp, low confidence
76Celeris-155.9%estimated ± 6.0 pp, low confidence
77Claude 3.5 Haiku55.9%estimated ± 6.0 pp, low confidence
78Command A+55.9%estimated ± 6.0 pp, low confidence
79DeepSeek V355.9%estimated ± 6.0 pp, low confidence
80DeepSeek V3 032455.9%estimated ± 6.0 pp, low confidence
81Gemini 2.5 Pro55.9%estimated ± 6.0 pp, low confidence
82Gemma 3 27B55.9%estimated ± 6.0 pp, low confidence
83Gemma 4 12B55.9%estimated ± 6.0 pp, low confidence
84Gemma 4 26B A4B55.9%estimated ± 6.0 pp, low confidence
85Gemma 4 31B55.9%estimated ± 6.0 pp, low confidence
86Gemma 4 E2B55.9%estimated ± 6.0 pp, low confidence
87Gemma 4 E4B55.9%estimated ± 6.0 pp, low confidence
88GPT-4.1 mini55.9%estimated ± 6.0 pp, low confidence
89GPT-4.1 nano55.9%estimated ± 6.0 pp, low confidence
90GPT-4o55.9%estimated ± 6.0 pp, low confidence
91GPT-4o mini55.9%estimated ± 6.0 pp, low confidence
92GPT-OSS 120B55.9%estimated ± 6.0 pp, low confidence
93GPT-OSS 20B55.9%estimated ± 6.0 pp, low confidence
94Granite 4.2 30B55.9%estimated ± 6.0 pp, low confidence
95Granite 4.2 3B55.9%estimated ± 6.0 pp, low confidence
96Granite 4.2 8B55.9%estimated ± 6.0 pp, low confidence
97K-Exaone55.9%estimated ± 6.0 pp, low confidence
98LFM2.5-2.6B55.9%estimated ± 6.0 pp, low confidence
99Ling 2.6 Flash55.9%estimated ± 6.0 pp, low confidence
100Ling 3.0 Tiny55.9%estimated ± 6.0 pp, low confidence
101Llama 3.3 70B55.9%estimated ± 6.0 pp, low confidence
102Llama 4 Maverick55.9%estimated ± 6.0 pp, low confidence
103Llama 4 Scout55.9%estimated ± 6.0 pp, low confidence
104Mercury 2.555.9%estimated ± 6.0 pp, low confidence
105MiMo-V2-Flash55.9%estimated ± 6.0 pp, low confidence
106Mistral Large 355.9%estimated ± 6.0 pp, low confidence
107Mistral Small 3.155.9%estimated ± 6.0 pp, low confidence
108Mistral Small 455.9%estimated ± 6.0 pp, low confidence
109Nemotron 3.5 Lightning 30B A3B NVFP455.9%estimated ± 6.0 pp, low confidence
110Nemotron 3 Nano 30B55.9%estimated ± 6.0 pp, low confidence
111Nemotron 3 Nano Omni 30B A3B55.9%estimated ± 6.0 pp, low confidence
112North Mini Code55.9%estimated ± 6.0 pp, low confidence
113Solar Pro 355.9%estimated ± 6.0 pp, low confidence
114Trinity-Large-Preview55.9%estimated ± 6.0 pp, low confidence
115Trinity-Large-Thinking55.9%estimated ± 6.0 pp, low confidence
116Ultravox v0.6 Llama 3.3 70B55.9%estimated ± 6.0 pp, low confidence
117Solar Open 255.4%estimated ± 2.6 pp, high confidence
118GPT-5.4 mini55.2%estimated ± 2.6 pp, high confidence
119GPT-5.4 nano54.9%estimated ± 2.6 pp, high confidence
120Kimi K2.654.9%estimated ± 2.6 pp, high confidence
121Nemotron 3.5 Bolt 74B-A9B BF1654.8%estimated ± 2.6 pp, high confidence
122Ornith-1.5-9B54.6%estimated ± 2.6 pp, high confidence
123Kimi K2.554.3%measured
124Nemotron 3.5 Bolt 74B-A9B54.3%estimated ± 2.6 pp, high confidence
125GLM-554.1%measured
126LLaDA2.2-flash53.7%estimated ± 2.6 pp, high confidence
127Qwen3.6-27B53.4%measured
128LFM3-1.7B53.2%estimated ± 2.6 pp, high confidence
129Qwen3.6-35B-A3B52.6%measured
130Claude Opus 4.552.3%measured
131Agents-A1-4B52.0%estimated ± 2.0 pp, medium confidence
132Agents-A152.0%estimated ± 2.0 pp, medium confidence
133Claude Sonnet 4.552.0%estimated ± 2.0 pp, low confidence
134DeepSeek V3.252.0%estimated ± 2.0 pp, low confidence
135GLM-4.752.0%estimated ± 2.0 pp, low confidence
136LongCat-Flash-Lite-Sparse52.0%estimated ± 2.0 pp, low confidence
137Qwen3.5 397B51.8%measured

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General