Model comparison
Mixtral 8x22B vs o1
o1 is the stronger model overall, scoring 40.9 to 27.1 on the Noometry Index. Mixtral 8x22B costs 8.8× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Mistral AISummary
- They share 25 benchmarks with published results for both. Mixtral 8x22B scores higher in 0 categories and o1 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o1 leads 41.5 to 15.1.
- The biggest single-benchmark swing is MATH Level 5: 24.2% for Mixtral 8x22B and 94.7% for o1.
- Mixtral 8x22B is cheaper at $2 / $6 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 64K.
- Mixtral 8x22B has downloadable open weights; the other is API-only.
Side by side
| Mixtral 8x22B | o1 | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 27.1 | 40.9 |
| Released | 2024-04-17 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 64K | 200K |
| Max output | 64K | 100K |
| Input $ / M tokens | $2 | $15 |
| Output $ / M tokens | $6 | $60 |
| Results tracked | 34 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
Mixtral 8x22B: 24.2 (#329), o1: 46.1 (#70)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| WeirdML | 3.2% | 47.6% |
| LMArena Coding | 1166 | 1367 |
| HumanEval+ | 72% | 89% |
| MBPP+ | 64.3% | 80.2% |
| Aider Polyglot | — | 61.7% |
| BigCodeBench Instruct | 40.6% | — |
| LiveBench Coding | — | 69.7% |
| BigCodeBench Complete | 50.2% | — |
| CadEval | — | 56% |
Agentic & Tool Use o1 leads
Mixtral 8x22B: 23.1 (#127), o1: 24.6 (#117)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| Cybench | 7.5% | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning o1 leads
Mixtral 8x22B: 19.9 (#248), o1: 27.9 (#111)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1371 |
| DTBench | 55.1% | 74.7% |
| Epoch Capabilities Index | 122.03 | 141.91 |
| SimpleBench | — | 41.7% |
| ARC-AGI-1 | — | 30.7% |
| Chess Puzzles | — | 15% |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| LiveBench Data Analysis | — | 65.5% |
| LMCA | — | 22.3% |
| ForecastBench | 56.3 | — |
| LiveBench | — | 75.7% |
Math o1 leads
Mixtral 8x22B: 22.9 (#275), o1: 36.1 (#175)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Math | 1184 | 1388 |
| MATH Level 5 | 24.2% | 94.7% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| Omni-MATH | 16.3% | — |
| LiveBench Math | — | 80.3% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge o1 leads
Mixtral 8x22B: 15.1 (#293), o1: 41.5 (#110)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| GPQA Diamond | 34.1% | 76.8% |
| LMArena Expert | 1113 | 1361 |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| MMLU-Pro | 46% | — |
| Confabulations | — | 11.7% |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |
Multimodal Not comparable
Mixtral 8x22B: —, o1: 34.2 (#93)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual o1 leads
Mixtral 8x22B: 32.8 (#255), o1: 48.6 (#142)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Non-English | 1128 | 1358 |
| LMArena Chinese | 1116 | 1394 |
| LMArena French | 1166 | 1344 |
| LMArena German | 1141 | 1337 |
| LMArena Japanese | 1037 | 1346 |
| LMArena Korean | 1057 | 1396 |
| LMArena Russian | 1158 | 1356 |
| LMArena Spanish | 1151 | 1345 |
Instruction Following o1 leads
Mixtral 8x22B: 57.7 (#266), o1: 74.8 (#86)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Instruction Following | 1147 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | 72.4% | — |
Long Context o1 leads
Mixtral 8x22B: 34.7 (#247), o1: 50.3 (#9)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Longer Query | 1144 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference o1 leads
Mixtral 8x22B: 36.9 (#262), o1: 55.6 (#144)
| Benchmark | Mixtral 8x22B | o1 |
|---|---|---|
| LMArena Text | 1162 | 1366 |
| LMArena Creative Writing | 1141 | 1348 |
| LMArena Multi-Turn | 1130 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| WildBench | 71.1% | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is Mixtral 8x22B better than o1?
o1 is the stronger model overall, scoring 40.9 to 27.1 on the Noometry Index. Mixtral 8x22B costs 8.8× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Which is cheaper, Mixtral 8x22B or o1?
Mixtral 8x22B is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; o1 lists at $15 and $60.
Is Mixtral 8x22B or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
o1 does, with 200K tokens against 64K.
How many benchmarks do Mixtral 8x22B and o1 share?
25 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and o1 has 52.