Model comparison
C4ai Aya Expanse 32b vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 35.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
CohereSummary
- They share 18 benchmarks with published results for both. C4ai Aya Expanse 32b scores higher in 0 categories and Kimi K2 (Jul 2025) in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 42.2.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 10.9% for C4ai Aya Expanse 32b and 17.9% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 128K.
Side by side
| C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Cohere | Moonshot AI |
| Noometry Index | 35.9 | 41.2 |
| Released | 2024-10-24 | 2025-07-12 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 262K |
| Input $ / M tokens | — | $0.57 |
| Output $ / M tokens | — | $2.30 |
| Results tracked | 18 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 34.8 (#231), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Coding | 1197 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| GSO | — | 4.9% |
| WeirdML | — | 42.8% |
| ALE-Bench | — | 597.5 |
Agentic & Tool Use Not comparable
C4ai Aya Expanse 32b: —, Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning Too close to call
C4ai Aya Expanse 32b: 23.3 (#180), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Hard Prompts | 1193 | 1384 |
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| Epoch Capabilities Index | — | 146.01 |
| ForecastBench | — | 60.2 |
Math Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 34.0 (#197), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1200 | 1397 |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 33.2 (#206), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| Vectara Hallucination Rate | 10.9% | 17.9% |
| LMArena Expert | 1182 | 1365 |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| GPQA (HELM) | — | 65.3% |
Multilingual Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 38.4 (#230), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1213 | 1372 |
| LMArena Chinese | 1211 | 1415 |
| LMArena French | 1248 | 1379 |
| LMArena German | 1199 | 1387 |
| LMArena Japanese | 1163 | 1349 |
| LMArena Korean | 1158 | 1325 |
| LMArena Russian | 1227 | 1385 |
| LMArena Spanish | 1193 | 1386 |
Instruction Following Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 62.6 (#237), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1196 | 1348 |
| IFEval | — | 85% |
Long Context Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 37.2 (#220), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1228 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
C4ai Aya Expanse 32b: 42.2 (#235), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | C4ai Aya Expanse 32b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1224 | 1380 |
| LMArena Creative Writing | 1200 | 1350 |
| LMArena Multi-Turn | 1190 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |
Frequently asked questions
Is C4ai Aya Expanse 32b better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 35.9 on the Noometry Index.
Is C4ai Aya Expanse 32b or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 34.8 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 128K.
How many benchmarks do C4ai Aya Expanse 32b and Kimi K2 (Jul 2025) share?
18 benchmarks have published results for both models. C4ai Aya Expanse 32b has 18 scored results on Noometry and Kimi K2 (Jul 2025) has 42.