Model comparison
C4ai Aya Expanse 8b vs Deepseek Coder v2
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 34.9 on the Noometry Index.
Last verified . 14 shared benchmarks.
Cohere
DeepSeekSummary
- They share 14 benchmarks with published results for both. C4ai Aya Expanse 8b scores higher in 2 categories and Deepseek Coder v2 in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where Deepseek Coder v2 leads 38.1 to 33.8.
Side by side
| C4ai Aya Expanse 8b | Deepseek Coder v2 | |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 34.9 | 35.9 |
| Released | — | 2024-06-17 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
C4ai Aya Expanse 8b: 33.8 (#252), Deepseek Coder v2: 38.1 (#183)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Coding | 1160 | 1251 |
| BigCodeBench Instruct | — | 48.2% |
| BigCodeBench Complete | — | 59.7% |
| HumanEval+ | — | 82.3% |
| MBPP+ | — | 75.1% |
Reasoning Deepseek Coder v2 leads
C4ai Aya Expanse 8b: 22.4 (#197), Deepseek Coder v2: 23.6 (#176)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Hard Prompts | 1156 | 1207 |
| WinoGrande | — | 83.7% |
Math Deepseek Coder v2 leads
C4ai Aya Expanse 8b: 33.3 (#203), Deepseek Coder v2: 34.9 (#190)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Math | 1168 | 1241 |
| GSM8K | — | 94.5% |
Knowledge C4ai Aya Expanse 8b leads
C4ai Aya Expanse 8b: 33.6 (#201), Deepseek Coder v2: 32.3 (#212)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Expert | 1153 | 1181 |
| Vectara Hallucination Rate | 9.5% | — |
| ARC (AI2) Challenge | — | 64.3% |
Multilingual Too close to call
C4ai Aya Expanse 8b: 36.1 (#242), Deepseek Coder v2: 36.3 (#240)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Non-English | 1180 | 1182 |
| LMArena Chinese | 1181 | 1201 |
| LMArena German | 1188 | 1164 |
| LMArena Japanese | 1120 | 1126 |
| LMArena Russian | 1197 | 1188 |
| LMArena French | — | 1185 |
| LMArena Korean | — | 1104 |
| LMArena Spanish | — | 1153 |
Instruction Following Deepseek Coder v2 leads
C4ai Aya Expanse 8b: 60.1 (#253), Deepseek Coder v2: 61.7 (#242)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Instruction Following | 1155 | 1180 |
Long Context Too close to call
C4ai Aya Expanse 8b: 36.1 (#236), Deepseek Coder v2: 37.0 (#224)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Longer Query | 1191 | 1219 |
Writing & Preference Too close to call
C4ai Aya Expanse 8b: 39.0 (#250), Deepseek Coder v2: 38.2 (#253)
| Benchmark | C4ai Aya Expanse 8b | Deepseek Coder v2 |
|---|---|---|
| LMArena Text | 1185 | 1191 |
| LMArena Creative Writing | 1168 | 1120 |
| LMArena Multi-Turn | 1160 | 1177 |
Frequently asked questions
Is C4ai Aya Expanse 8b better than Deepseek Coder v2?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 34.9 on the Noometry Index.
Is C4ai Aya Expanse 8b or Deepseek Coder v2 better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 33.8 in the Noometry coding category.
How many benchmarks do C4ai Aya Expanse 8b and Deepseek Coder v2 share?
14 benchmarks have published results for both models. C4ai Aya Expanse 8b has 15 scored results on Noometry and Deepseek Coder v2 has 24.