Model comparison
DeepSeek LLM 67B vs gpt-oss-20b
gpt-oss-20b is the stronger model overall, scoring 32.1 to 24.8 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeekSummary
- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and gpt-oss-20b in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-20b leads 37.6 to 8.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 65.3% for gpt-oss-20b.
Side by side
| DeepSeek LLM 67B | gpt-oss-20b | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 24.8 | 32.1 |
| Released | 2023-11-29 | 2025-08-05 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.018 |
| Output $ / M tokens | — | $0.09 |
| Results tracked | 15 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
DeepSeek LLM 67B: 31.3 (#273), gpt-oss-20b: 37.1 (#185)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1096 | 1307 |
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| ALE-Bench | — | 566.05 |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, gpt-oss-20b: 9.2 (#146)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
Reasoning gpt-oss-20b leads
DeepSeek LLM 67B: 15.9 (#300), gpt-oss-20b: 18.8 (#251)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| Chess Puzzles | 0% | 4% |
| LMArena Hard Prompts | 1070 | 1274 |
| Epoch Capabilities Index | 110.54 | 137.81 |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
Math gpt-oss-20b leads
DeepSeek LLM 67B: 8.1 (#318), gpt-oss-20b: 37.6 (#93)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 65.3% |
| LMArena Math | 1108 | 1317 |
| Omni-MATH | — | 56.5% |
| MATH Level 5 | 6.4% | — |
Knowledge gpt-oss-20b leads
DeepSeek LLM 67B: 7.1 (#308), gpt-oss-20b: 34.6 (#186)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 24.6% | 60.8% |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |
| LMArena Expert | — | 1258 |
Multilingual gpt-oss-20b leads
DeepSeek LLM 67B: 29.5 (#262), gpt-oss-20b: 42.4 (#192)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1073 | 1269 |
| LMArena Chinese | 1132 | 1311 |
| LMArena German | — | 1249 |
| LMArena Japanese | — | 1245 |
| LMArena Korean | — | 1239 |
| LMArena Russian | — | 1279 |
| LMArena Spanish | — | 1267 |
Instruction Following gpt-oss-20b leads
DeepSeek LLM 67B: 55.7 (#272), gpt-oss-20b: 62.0 (#235)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1079 | 1236 |
| IFEval | — | 73.2% |
Long Context gpt-oss-20b leads
DeepSeek LLM 67B: 33.1 (#260), gpt-oss-20b: 37.9 (#203)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1092 | 1251 |
Writing & Preference gpt-oss-20b leads
DeepSeek LLM 67B: 31.8 (#277), gpt-oss-20b: 35.8 (#260)
| Benchmark | DeepSeek LLM 67B | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1105 | 1288 |
| LMArena Creative Writing | 1067 | 1201 |
| LMArena Multi-Turn | 1082 | 1268 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
Frequently asked questions
Is DeepSeek LLM 67B better than gpt-oss-20b?
gpt-oss-20b is the stronger model overall, scoring 32.1 to 24.8 on the Noometry Index.
Is DeepSeek LLM 67B or gpt-oss-20b better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.1 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and gpt-oss-20b share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and gpt-oss-20b has 34.