Model comparison
gpt-oss-120b vs Mixtral 8x22B
gpt-oss-120b is the stronger model overall, scoring 36.3 to 27.1 on the Noometry Index.
Last verified . 26 shared benchmarks.
Mistral AISummary
- They share 26 benchmarks with published results for both. gpt-oss-120b scores higher in 7 categories and Mixtral 8x22B in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 22.9.
- The biggest single-benchmark swing is Omni-MATH: 68.8% for gpt-oss-120b and 16.3% for Mixtral 8x22B.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- gpt-oss-120b accepts more context: 131K tokens versus 64K.
Side by side
| gpt-oss-120b | Mixtral 8x22B | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 36.3 | 27.1 |
| Released | 2025-08-05 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 131K | 64K |
| Max output | 41K | 64K |
| Input $ / M tokens | $0.037 | $2 |
| Output $ / M tokens | $0.17 | $6 |
| Results tracked | 48 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-120b leads
gpt-oss-120b: 33.5 (#256), Mixtral 8x22B: 24.2 (#329)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| WeirdML | 48.2% | 3.2% |
| LMArena Coding | 1380 | 1166 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| SciCode | 36% | — |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Mixtral 8x22B leads
gpt-oss-120b: 12.2 (#153), Mixtral 8x22B: 23.1 (#127)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| Cybench | — | 7.5% |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Too close to call
gpt-oss-120b: 20.0 (#245), Mixtral 8x22B: 19.9 (#248)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1364 | 1150 |
| DTBench | 76.3% | 55.1% |
| Epoch Capabilities Index | 139.93 | 122.03 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| Mystery Game Puzzles | 2% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| ForecastBench | — | 56.3 |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Mixtral 8x22B: 22.9 (#275)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| Omni-MATH | 68.8% | 16.3% |
| LMArena Math | 1389 | 1184 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| MATH Level 5 | — | 24.2% |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), Mixtral 8x22B: 15.1 (#293)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 75.8% | 34.1% |
| MMLU-Pro | 79.5% | 46% |
| GPQA (HELM) | 68.4% | 33.4% |
| LMArena Expert | 1356 | 1113 |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| MMLU | — | 77.8% |
Multilingual gpt-oss-120b leads
gpt-oss-120b: 48.0 (#147), Mixtral 8x22B: 32.8 (#255)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1351 | 1128 |
| LMArena Chinese | 1385 | 1116 |
| LMArena French | 1369 | 1166 |
| LMArena German | 1353 | 1141 |
| LMArena Japanese | 1331 | 1037 |
| LMArena Korean | 1282 | 1057 |
| LMArena Russian | 1343 | 1158 |
| LMArena Spanish | 1389 | 1151 |
Instruction Following gpt-oss-120b leads
gpt-oss-120b: 69.3 (#173), Mixtral 8x22B: 57.7 (#266)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| IFEval | 83.6% | 72.4% |
| LMArena Instruction Following | 1318 | 1147 |
Long Context Mixtral 8x22B leads
gpt-oss-120b: 31.4 (#278), Mixtral 8x22B: 34.7 (#247)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1319 | 1144 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference gpt-oss-120b leads
gpt-oss-120b: 46.5 (#217), Mixtral 8x22B: 36.9 (#262)
| Benchmark | gpt-oss-120b | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1365 | 1162 |
| LMArena Creative Writing | 1275 | 1141 |
| WildBench | 84.5% | 71.1% |
| LMArena Multi-Turn | 1340 | 1130 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
Frequently asked questions
Is gpt-oss-120b better than Mixtral 8x22B?
gpt-oss-120b is the stronger model overall, scoring 36.3 to 27.1 on the Noometry Index.
Which is cheaper, gpt-oss-120b or Mixtral 8x22B?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is gpt-oss-120b or Mixtral 8x22B better for coding?
gpt-oss-120b scores higher on coding benchmarks: 33.5 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
gpt-oss-120b does, with 131K tokens against 64K.
How many benchmarks do gpt-oss-120b and Mixtral 8x22B share?
26 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Mixtral 8x22B has 34.