Model comparison
C4ai Aya Expanse 32b vs gpt-oss-20b
C4ai Aya Expanse 32b is the stronger model overall, scoring 35.9 to 32.5 on the Noometry Index.
Last verified . 16 shared benchmarks.
CohereSummary
- They share 16 benchmarks with published results for both. C4ai Aya Expanse 32b scores higher in 3 categories and gpt-oss-20b in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where C4ai Aya Expanse 32b leads 42.2 to 35.5.
- gpt-oss-20b accepts more context: 131K tokens versus 128K.
Side by side
| C4ai Aya Expanse 32b | gpt-oss-20b | |
|---|---|---|
| Provider | Cohere | OpenAI |
| Noometry Index | 35.9 | 32.5 |
| Released | 2024-10-24 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 16K |
| Input $ / M tokens | — | $0.018 |
| Output $ / M tokens | — | $0.09 |
| Results tracked | 18 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
C4ai Aya Expanse 32b: 34.8 (#231), gpt-oss-20b: 37.6 (#192)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1197 | 1306 |
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| ALE-Bench | — | 566.05 |
Agentic & Tool Use Not comparable
C4ai Aya Expanse 32b: —, gpt-oss-20b: 9.3 (#154)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
Reasoning C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 23.3 (#180), gpt-oss-20b: 19.3 (#261)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1193 | 1274 |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Epoch Capabilities Index | — | 137.82 |
Math gpt-oss-20b leads
C4ai Aya Expanse 32b: 34.0 (#197), gpt-oss-20b: 39.4 (#103)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Math | 1200 | 1317 |
| OTIS Mock AIME 2024-2025 | — | 65.3% |
| Omni-MATH | — | 56.5% |
Knowledge gpt-oss-20b leads
C4ai Aya Expanse 32b: 33.2 (#206), gpt-oss-20b: 34.6 (#195)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Expert | 1182 | 1258 |
| GPQA Diamond | — | 60.8% |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 10.9% | — |
| GPQA (HELM) | — | 59.4% |
Multilingual gpt-oss-20b leads
C4ai Aya Expanse 32b: 38.4 (#230), gpt-oss-20b: 42.2 (#197)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1213 | 1268 |
| LMArena Chinese | 1211 | 1314 |
| LMArena German | 1199 | 1255 |
| LMArena Japanese | 1163 | 1244 |
| LMArena Korean | 1158 | 1236 |
| LMArena Russian | 1227 | 1278 |
| LMArena Spanish | 1193 | 1267 |
| LMArena French | 1248 | — |
Instruction Following Too close to call
C4ai Aya Expanse 32b: 62.6 (#237), gpt-oss-20b: 61.8 (#240)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1196 | 1236 |
| IFEval | — | 73.2% |
Long Context Too close to call
C4ai Aya Expanse 32b: 37.2 (#220), gpt-oss-20b: 37.9 (#209)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1228 | 1250 |
Writing & Preference C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 42.2 (#235), gpt-oss-20b: 35.5 (#265)
| Benchmark | C4ai Aya Expanse 32b | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1224 | 1287 |
| LMArena Creative Writing | 1200 | 1201 |
| LMArena Multi-Turn | 1190 | 1268 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
Frequently asked questions
Is C4ai Aya Expanse 32b better than gpt-oss-20b?
C4ai Aya Expanse 32b is the stronger model overall, scoring 35.9 to 32.5 on the Noometry Index.
Is C4ai Aya Expanse 32b or gpt-oss-20b better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 34.8 in the Noometry coding category.
Which has the bigger context window?
gpt-oss-20b does, with 131K tokens against 128K.
How many benchmarks do C4ai Aya Expanse 32b and gpt-oss-20b share?
16 benchmarks have published results for both models. C4ai Aya Expanse 32b has 18 scored results on Noometry and gpt-oss-20b has 34.