Model comparison
DeepSeek-V3.2-Exp vs gpt-oss-120b
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 36.0 on the Noometry Index. gpt-oss-120b costs 4.1× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
DeepSeekSummary
- They share 34 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 8 categories and gpt-oss-120b in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where DeepSeek-V3.2-Exp leads 37.9 to 12.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 70% for DeepSeek-V3.2-Exp and 26% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
- DeepSeek-V3.2-Exp accepts more context: 164K tokens versus 131K.
Side by side
| DeepSeek-V3.2-Exp | gpt-oss-120b | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 44.9 | 36.0 |
| Released | 2025-09-29 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 164K | 131K |
| Max output | 66K | 41K |
| Input $ / M tokens | $0.26 | $0.037 |
| Output $ / M tokens | $0.38 | $0.17 |
| Results tracked | 48 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 46.4 (#61), gpt-oss-120b: 33.2 (#251)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| SWE-bench Verified (bash only) | 70% | 26% |
| Aider Polyglot | 74.2% | 41.8% |
| SciCode | 38.9% | 36% |
| WeirdML | 39.5% | 48.2% |
| LMArena Coding | 1454 | 1380 |
| LMArena WebDev | 1362 | — |
| SWE-bench Multilingual | 59% | — |
| ALE-Bench | — | 575.62 |
| AlgoTune | — | 1.41 |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 37.9 (#31), gpt-oss-120b: 12.1 (#145)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | 39.6% | 18.7% |
| Vending-Bench 2 | 1,034 | -21.53 |
| APEX-Agents | — | 4.4% |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| METR Time Horizons | — | 56.6% |
Reasoning DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 22.0 (#185), gpt-oss-120b: 19.5 (#236)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| Kagi LLM Benchmark | 52.2% | 58.6% |
| CritPt | 2.9% | 1.1% |
| Chess Puzzles | 14% | 20% |
| LMArena Hard Prompts | 1433 | 1364 |
| DTBench | 87.7% | 76.3% |
| LMCA | 29.1% | 22.1% |
| Epoch Capabilities Index | 146.28 | 139.94 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 22.1% |
| NYT Connections (extended) | 36.7% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
| Mystery Game Puzzles | — | 2% |
| Surface Evolver Bench | — | 25% |
Math gpt-oss-120b leads
DeepSeek-V3.2-Exp: 40.8 (#75), gpt-oss-120b: 50.6 (#43)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 88.9% |
| LMArena Math | 1435 | 1388 |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| Omni-MATH | — | 68.8% |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.6 (#61), gpt-oss-120b: 42.3 (#89)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 83.4% | 75.8% |
| Vectara Hallucination Rate | 5.3% | 14.2% |
| LMArena Expert | 1436 | 1356 |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| GPQA (HELM) | — | 68.4% |
Multilingual DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 52.4 (#85), gpt-oss-120b: 48.2 (#142)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1409 | 1350 |
| LMArena Chinese | 1461 | 1384 |
| LMArena French | 1435 | 1370 |
| LMArena German | 1440 | 1355 |
| LMArena Japanese | 1377 | 1332 |
| LMArena Korean | 1372 | 1283 |
| LMArena Russian | 1423 | 1343 |
| LMArena Spanish | 1441 | 1388 |
Instruction Following DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 74.7 (#89), gpt-oss-120b: 69.5 (#168)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1413 | 1318 |
| IFEval | — | 83.6% |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), gpt-oss-120b: 31.4 (#273)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| Fiction.LiveBench | 83.3% | 44.4% |
| LMArena Longer Query | 1428 | 1319 |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 62.5 (#73), gpt-oss-120b: 46.7 (#213)
| Benchmark | DeepSeek-V3.2-Exp | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1425 | 1366 |
| LMArena Creative Writing | 1403 | 1274 |
| EQ-Bench Creative Writing | 1515 | 961 |
| LMArena Multi-Turn | 1427 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| WildBench | — | 84.5% |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than gpt-oss-120b?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 36.0 on the Noometry Index. gpt-oss-120b costs 4.1× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.2-Exp or gpt-oss-120b?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.
Is DeepSeek-V3.2-Exp or gpt-oss-120b better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.2-Exp does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-V3.2-Exp and gpt-oss-120b share?
34 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and gpt-oss-120b has 48.