aya-expanse-32b
Cohere Labs- Arena score
- 1224
- Among open models
- #107of 148
- Among all on Arena
- #314
- Votes
- 27,124
Measured on Aya Expanse 32B, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- French1248#78#245274
- Legal & government1248#100#2941.7K
- Science1233#103#3054.8K
- English1229#109#31714.4K
- Long prompts1227#104#3054.2K
- Russian1227#101#2993.9K
- Medicine & health1222#99#2911.4K
- Writing & literature1222#103#3077.3K
- Non-English1213#103#30512.8K
- Software & IT1212#109#3197.3K
- Chinese1211#103#3062.4K
- Business & finance1210#103#3083.1K
- Applied math1208#105#3073.2K
- Math1200#105#3103.9K
- Creative writing1200#106#3123.9K
- German1198#83#261609
- Coding1196#109#3214.7K
- Instructions1196#108#31711.3K
- Hard prompts, English1196#110#3224.3K
- Hard prompts1193#110#3207.2K
- Spanish1193#89#266256
- Multi-turn1190#110#3205.2K
- Media & sports1183#107#3164.3K
- Expert1182#103#2991.8K
- Japanese1163#67#233670
- Korean1158#74#242385
- Excluding ties1115#108#31517.1K
About as strong, smaller
Compare
granite-4.2-3bIBM Granite12923.7B
gemma-3-4b-itGoogle12914.3B

gemma-3n-E4B-itGoogle13067.8B

Molmo2-8BAi212878.7B
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiobartowski/aya-expanse-32b-GGUF16 builds
- MLX · Apple silicon: mlx-lm, LM Studiomlx-community/aya-expanse-32b-8bit1 build
- AWQ · vLLM, SGLangOrion-zhen/aya-expanse-32b-AWQ1 build
- GPTQ · vLLM, SGLangkevinbazira/aya-expanse-32b-gptq-4bit1 build
Most downloaded
- aya-expanse-32b-GGUFbartowski · Quantization1.6K downloads
- aya-expanse-32b-gptq-4bitkevinbazira · Quantization883 downloads
- aya-expanse-32b-AWQOrion-zhen · Quantization445 downloads
- aya-expanse-32b-GGUFlmstudio-community · Quantization205 downloads
- aya-expanse-32b-GGUFtensorblock · Quantization149 downloads
BenchClub runs
Not runnable yet
Runs open once a aya-expanse-32b build is registered.
More from Cohere Labs
See more
tiny-aya-global31.7K3.3B

aya-expanse-8b26.3K8.0B

c4ai-command-r-plus202104B
Other 20–40B models
See more
Qwen3.8-27BQwen6.9M27.8B

gemma-4-31B-itGoogle6.3M31.3B
Ornith-1.0-35BOrnith1.4M35.1B
NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1.3M31.6B