Qwen2-72B-Instruct
Qwen- Arena score
- 1203
- Among open models
- #111of 148
- Among all on Arena
- #324
- Votes
- 37,325
Measured on Qwen2-72B-Instruct, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Chinese1240#95#2844.1K
- Math1235#100#2904.8K
- Applied math1231#100#2954.2K
- English1222#110#32219.9K
- Hard prompts, English1206#108#3166.5K
- Writing & literature1202#108#32110K
- Coding1196#110#3226.2K
- Multi-turn1195#108#3176.7K
- Science1195#111#3256.5K
- Legal & government1193#110#3212.3K
- Software & IT1193#113#32810.3K
- Long prompts1192#110#3244.6K
- Hard prompts1191#111#32310.8K
- Instructions1181#110#32314.2K
- Creative writing1181#110#3235.5K
- Non-English1176#112#32617.4K
- Medicine & health1176#109#3172.1K
- Expert1171#107#3071.8K
- French1170#87#265596
- Spanish1169#92#271596
- Russian1169#112#3284.6K
- Media & sports1166#108#3215.8K
- German1151#89#2801.3K
- Business & finance1146#114#3314.3K
- Japanese1111#75#252901
- Excluding ties1087#111#32223.8K
- Korean1083#81#2611.2K
About as strong, smaller
Compare
granite-4.2-3bIBM Granite12923.7B
gemma-3-4b-itGoogle12914.3B

gemma-3n-E4B-itGoogle13067.8B

Llama-3.1-Tulu-3-8BAi211938.0B
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiolegraphista/Qwen2-72B-Instruct-IMat-GGUF13 builds
- AWQ · vLLM, SGLangQwen/Qwen2-72B-Instruct-AWQ1 build
- GPTQ · vLLM, SGLangQwen/Qwen2-72B-Instruct-GPTQ-Int43 builds
Most downloaded
- Qwen2-72B-Instruct-IMat-GGUFlegraphista · Quantization1.2K downloads
- Qwen2-72B-Instruct-GGUFsecond-state · Quantization853 downloads
- Qwen2-72B-Instruct-iMat-GGUFqwp4w3hyb · Quantization593 downloads
- Qwen2-72B-Instruct-i1-GGUFmradermacher · Quantization569 downloads
- Qwen2-72B-Instruct-GGUFmradermacher · Quantization422 downloads
BenchClub runs
Not runnable yet
Runs open once a Qwen2-72B-Instruct build is registered.
More from Qwen
See more
Qwen3.8-27B6.9M27.8B

Qwen3.5-27B1.7M27.8B

Qwen3.5-35B-A3B1.4M36.0B

Qwen3.5-122B-A10B775.1K125B
Other 40–80B models
See more
Llama-3.3-70B-InstructMeta Llama344.4K70.6B

Mixtral-8x7B-Instruct-v0.1Mistral AI325.3K46.7B

falcon-40bTII11.6K41.8B
Jamba-v0.1AI21 Labs4.5K51.6B