Model comparison
C4ai Aya Expanse 32b vs Mistral Small
C4ai Aya Expanse 32b is the stronger model overall, scoring 35.9 to 33.4 on the Noometry Index.
Last verified . 18 shared benchmarks.
Cohere
Mistral AISummary
- They share 18 benchmarks with published results for both. C4ai Aya Expanse 32b scores higher in 4 categories and Mistral Small in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where C4ai Aya Expanse 32b leads 34.0 to 16.4.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 10.9% for C4ai Aya Expanse 32b and 5.1% for Mistral Small.
- Mistral Small accepts more context: 262K tokens versus 128K.
Side by side
| C4ai Aya Expanse 32b | Mistral Small | |
|---|---|---|
| Provider | Cohere | Mistral AI |
| Noometry Index | 35.9 | 33.4 |
| Released | 2024-10-24 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 256K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 18 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
C4ai Aya Expanse 32b: 34.8 (#231), Mistral Small: 34.0 (#247)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Coding | 1197 | 1362 |
| SciCode | — | 26.5% |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
| ALE-Bench | — | 497.62 |
Agentic & Tool Use Not comparable
C4ai Aya Expanse 32b: —, Mistral Small: 28.1 (#93)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.1% |
Reasoning C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 23.3 (#180), Mistral Small: 19.8 (#250)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Hard Prompts | 1193 | 1335 |
| Kagi LLM Benchmark | — | 37.8% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 44.8% |
| DTBench | — | 70.9% |
| LiveBench Data Analysis | — | 53.7% |
| LMCA | — | 20.6% |
| LiveBench | — | 44% |
Math C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 34.0 (#197), Mistral Small: 16.4 (#293)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Math | 1200 | 1341 |
| OTIS Mock AIME 2024-2025 | — | 5.8% |
| LiveBench Math | — | 39.9% |
| MATH Level 5 | — | 46.8% |
Knowledge C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 33.2 (#206), Mistral Small: 31.0 (#222)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| Vectara Hallucination Rate | 10.9% | 5.1% |
| LMArena Expert | 1182 | 1291 |
| GPQA Diamond | — | 47.5% |
| MMLU | — | 68.7% |
Multimodal Not comparable
C4ai Aya Expanse 32b: —, Mistral Small: 33.5 (#96)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Vision | — | 1142 |
Multilingual Mistral Small leads
C4ai Aya Expanse 32b: 38.4 (#230), Mistral Small: 45.5 (#169)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Non-English | 1213 | 1315 |
| LMArena Chinese | 1211 | 1340 |
| LMArena French | 1248 | 1337 |
| LMArena German | 1199 | 1340 |
| LMArena Japanese | 1163 | 1275 |
| LMArena Korean | 1158 | 1259 |
| LMArena Russian | 1227 | 1324 |
| LMArena Spanish | 1193 | 1346 |
Instruction Following Mistral Small leads
C4ai Aya Expanse 32b: 62.6 (#237), Mistral Small: 66.4 (#209)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1196 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
Long Context Mistral Small leads
C4ai Aya Expanse 32b: 37.2 (#220), Mistral Small: 40.4 (#156)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1228 | 1327 |
Writing & Preference Mistral Small leads
C4ai Aya Expanse 32b: 42.2 (#235), Mistral Small: 52.5 (#171)
| Benchmark | C4ai Aya Expanse 32b | Mistral Small |
|---|---|---|
| LMArena Text | 1224 | 1338 |
| LMArena Creative Writing | 1200 | 1305 |
| LMArena Multi-Turn | 1190 | 1344 |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is C4ai Aya Expanse 32b better than Mistral Small?
C4ai Aya Expanse 32b is the stronger model overall, scoring 35.9 to 33.4 on the Noometry Index.
Is C4ai Aya Expanse 32b or Mistral Small better for coding?
They score almost the same on coding (34.8 vs 34.0); test both on your own repository before choosing.
Which has the bigger context window?
Mistral Small does, with 262K tokens against 128K.
How many benchmarks do C4ai Aya Expanse 32b and Mistral Small share?
18 benchmarks have published results for both models. C4ai Aya Expanse 32b has 18 scored results on Noometry and Mistral Small has 39.