Model comparison
Deepseek Coder v2 vs o1-mini
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 34.0 on the Noometry Index.
Last verified . 19 shared benchmarks.
DeepSeekSummary
- They share 19 benchmarks with published results for both. Deepseek Coder v2 scores higher in 2 categories and o1-mini in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Deepseek Coder v2 leads 23.6 to 8.8.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | o1-mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 35.9 | 34.0 |
| Released | 2024-06-17 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), o1-mini: 35.5 (#224)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Coding | 1251 | 1362 |
| HumanEval+ | 82.3% | 89% |
| MBPP+ | 75.1% | 78.8% |
| Aider Polyglot | — | 32.9% |
| WeirdML | — | 36.3% |
| BigCodeBench Instruct | 48.2% | — |
| LiveBench Coding | — | 48% |
| BigCodeBench Complete | 59.7% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, o1-mini: 24.6 (#118)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| Cybench | — | 10% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), o1-mini: 8.8 (#346)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1333 |
| ARC-AGI-2 | — | 0.8% |
| SimpleBench | — | 18.1% |
| ARC-AGI-1 | — | 14% |
| LiveBench Reasoning | — | 72.3% |
| LiveBench Data Analysis | — | 57.9% |
| Epoch Capabilities Index | — | 135.82 |
| LiveBench | — | 57.8% |
| WinoGrande | 83.7% | — |
Math Too close to call
Deepseek Coder v2: 34.9 (#190), o1-mini: 35.4 (#186)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Math | 1241 | 1358 |
| OTIS Mock AIME 2024-2025 | — | 46.9% |
| LiveBench Math | — | 62% |
| MATH Level 5 | — | 89.2% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
| GSM8K | 94.5% | — |
Knowledge o1-mini leads
Deepseek Coder v2: 32.3 (#212), o1-mini: 34.9 (#192)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Expert | 1181 | 1316 |
| GPQA Diamond | — | 62.4% |
| Confabulations | — | 18.6% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual o1-mini leads
Deepseek Coder v2: 36.3 (#240), o1-mini: 43.6 (#182)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Non-English | 1182 | 1289 |
| LMArena Chinese | 1201 | 1314 |
| LMArena French | 1185 | 1293 |
| LMArena German | 1164 | 1278 |
| LMArena Japanese | 1126 | 1245 |
| LMArena Korean | 1104 | 1223 |
| LMArena Russian | 1188 | 1283 |
| LMArena Spanish | 1153 | 1303 |
Instruction Following o1-mini leads
Deepseek Coder v2: 61.7 (#242), o1-mini: 66.7 (#206)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1180 | 1304 |
| LiveBench Instruction Following | — | 65.4% |
Long Context o1-mini leads
Deepseek Coder v2: 37.0 (#224), o1-mini: 40.1 (#161)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Longer Query | 1219 | 1320 |
Writing & Preference o1-mini leads
Deepseek Coder v2: 38.2 (#253), o1-mini: 48.4 (#202)
| Benchmark | Deepseek Coder v2 | o1-mini |
|---|---|---|
| LMArena Text | 1191 | 1317 |
| LMArena Creative Writing | 1120 | 1244 |
| LMArena Multi-Turn | 1177 | 1314 |
| Short-Story Creative Writing | — | 64.9% |
| LiveBench Language | — | 40.9% |
Frequently asked questions
Is Deepseek Coder v2 better than o1-mini?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 34.0 on the Noometry Index.
Is Deepseek Coder v2 or o1-mini better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 35.5 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and o1-mini share?
19 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and o1-mini has 39.