Model comparison
C4ai Aya Expanse 32b vs Gemma 3 4B
C4ai Aya Expanse 32b is the stronger model overall, scoring 35.9 to 28.1 on the Noometry Index.
Last verified . 13 shared benchmarks.
Cohere
GoogleSummary
- They share 13 benchmarks with published results for both. C4ai Aya Expanse 32b scores higher in 4 categories and Gemma 3 4B in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where C4ai Aya Expanse 32b leads 33.2 to 11.8.
- Gemma 3 4B accepts more context: 131K tokens versus 128K.
Side by side
| C4ai Aya Expanse 32b | Gemma 3 4B | |
|---|---|---|
| Provider | Cohere | |
| Noometry Index | 35.9 | 28.1 |
| Released | 2024-10-24 | 2025-03-12 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 4K |
| Input $ / M tokens | — | $0.04 |
| Output $ / M tokens | — | $0.08 |
| Results tracked | 18 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
C4ai Aya Expanse 32b: 34.8 (#231), Gemma 3 4B: 35.9 (#215)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Coding | 1197 | 1230 |
Agentic & Tool Use Not comparable
C4ai Aya Expanse 32b: —, Gemma 3 4B: 20.9 (#142)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 19.6% |
Reasoning C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 23.3 (#180), Gemma 3 4B: 13.2 (#335)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Hard Prompts | 1193 | 1253 |
| Kagi LLM Benchmark | — | 25.2% |
| Chess Puzzles | — | 0% |
| DTBench | — | 50.9% |
| LMCA | — | 2.8% |
| Epoch Capabilities Index | — | 116.02 |
Math C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 34.0 (#197), Gemma 3 4B: 16.8 (#292)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Math | 1200 | 1239 |
| OTIS Mock AIME 2024-2025 | — | 7.5% |
Knowledge C4ai Aya Expanse 32b leads
C4ai Aya Expanse 32b: 33.2 (#206), Gemma 3 4B: 11.8 (#299)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| Vectara Hallucination Rate | 10.9% | 6.4% |
| LMArena Expert | 1182 | 1223 |
| GPQA Diamond | — | 23.2% |
Multilingual Gemma 3 4B leads
C4ai Aya Expanse 32b: 38.4 (#230), Gemma 3 4B: 42.5 (#194)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Non-English | 1213 | 1273 |
| LMArena German | 1199 | 1281 |
| LMArena Russian | 1227 | 1294 |
| LMArena Chinese | 1211 | — |
| LMArena French | 1248 | — |
| LMArena Japanese | 1163 | — |
| LMArena Korean | 1158 | — |
| LMArena Spanish | 1193 | — |
Instruction Following Gemma 3 4B leads
C4ai Aya Expanse 32b: 62.6 (#237), Gemma 3 4B: 65.2 (#225)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Instruction Following | 1196 | 1239 |
Long Context Gemma 3 4B leads
C4ai Aya Expanse 32b: 37.2 (#220), Gemma 3 4B: 38.7 (#194)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Longer Query | 1228 | 1273 |
Writing & Preference Too close to call
C4ai Aya Expanse 32b: 42.2 (#235), Gemma 3 4B: 42.0 (#239)
| Benchmark | C4ai Aya Expanse 32b | Gemma 3 4B |
|---|---|---|
| LMArena Text | 1224 | 1291 |
| LMArena Creative Writing | 1200 | 1271 |
| LMArena Multi-Turn | 1190 | 1255 |
| EQ-Bench Creative Writing | — | 1068 |
Frequently asked questions
Is C4ai Aya Expanse 32b better than Gemma 3 4B?
C4ai Aya Expanse 32b is the stronger model overall, scoring 35.9 to 28.1 on the Noometry Index.
Is C4ai Aya Expanse 32b or Gemma 3 4B better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 34.8 in the Noometry coding category.
Which has the bigger context window?
Gemma 3 4B does, with 131K tokens against 128K.
How many benchmarks do C4ai Aya Expanse 32b and Gemma 3 4B share?
13 benchmarks have published results for both models. C4ai Aya Expanse 32b has 18 scored results on Noometry and Gemma 3 4B has 22.