Model comparison
Deepseek Coder v2 vs gpt-oss-20b
Deepseek Coder v2 is the stronger model overall, scoring 35.3 to 32.1 on the Noometry Index.
Last verified . 16 shared benchmarks.
DeepSeekSummary
- They share 16 benchmarks with published results for both. Deepseek Coder v2 scores higher in 3 categories and gpt-oss-20b in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where gpt-oss-20b leads 42.4 to 36.4.
Side by side
| Deepseek Coder v2 | gpt-oss-20b | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 35.3 | 32.1 |
| Released | 2024-06-17 | 2025-08-05 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.018 |
| Output $ / M tokens | — | $0.09 |
| Results tracked | 24 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Deepseek Coder v2: 37.5 (#178), gpt-oss-20b: 37.1 (#185)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1251 | 1307 |
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 566.05 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, gpt-oss-20b: 9.2 (#146)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 22.6 (#175), gpt-oss-20b: 18.8 (#251)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1274 |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Epoch Capabilities Index | — | 137.81 |
| WinoGrande | 83.7% | — |
Math gpt-oss-20b leads
Deepseek Coder v2: 32.3 (#188), gpt-oss-20b: 37.6 (#93)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Math | 1241 | 1317 |
| OTIS Mock AIME 2024-2025 | — | 65.3% |
| Omni-MATH | — | 56.5% |
| GSM8K | 94.5% | — |
Knowledge gpt-oss-20b leads
Deepseek Coder v2: 31.7 (#209), gpt-oss-20b: 34.6 (#186)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Expert | 1181 | 1258 |
| GPQA Diamond | — | 60.8% |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual gpt-oss-20b leads
Deepseek Coder v2: 36.4 (#235), gpt-oss-20b: 42.4 (#192)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1182 | 1269 |
| LMArena Chinese | 1201 | 1311 |
| LMArena German | 1164 | 1249 |
| LMArena Japanese | 1126 | 1245 |
| LMArena Korean | 1104 | 1239 |
| LMArena Russian | 1188 | 1279 |
| LMArena Spanish | 1153 | 1267 |
| LMArena French | 1184 | — |
Instruction Following Too close to call
Deepseek Coder v2: 61.9 (#237), gpt-oss-20b: 62.0 (#235)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1180 | 1236 |
| IFEval | — | 73.2% |
Long Context Too close to call
Deepseek Coder v2: 37.0 (#219), gpt-oss-20b: 37.9 (#203)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1219 | 1251 |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.4 (#249), gpt-oss-20b: 35.8 (#260)
| Benchmark | Deepseek Coder v2 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1191 | 1288 |
| LMArena Creative Writing | 1120 | 1201 |
| LMArena Multi-Turn | 1176 | 1268 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
Frequently asked questions
Is Deepseek Coder v2 better than gpt-oss-20b?
Deepseek Coder v2 is the stronger model overall, scoring 35.3 to 32.1 on the Noometry Index.
Is Deepseek Coder v2 or gpt-oss-20b better for coding?
They score almost the same on coding (37.5 vs 37.1); test both on your own repository before choosing.
How many benchmarks do Deepseek Coder v2 and gpt-oss-20b share?
16 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and gpt-oss-20b has 34.