Model comparison
Mistral Medium vs Mixtral 8x22B
Mistral Medium is the stronger model overall, scoring 36.3 to 27.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Mistral AISummary
- They share 21 benchmarks with published results for both. Mistral Medium scores higher in 9 categories and Mixtral 8x22B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 36.9.
- The biggest single-benchmark swing is MATH Level 5: 81.6% for Mistral Medium and 24.2% for Mixtral 8x22B.
- Both cost about the same: $1.50 input and $7.50 output per million tokens.
- Mistral Medium accepts more context: 262K tokens versus 64K.
Side by side
| Mistral Medium | Mixtral 8x22B | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 36.3 | 27.1 |
| Released | 2023-12-11 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 262K | 64K |
| Max output | 262K | 64K |
| Input $ / M tokens | $1.50 | $2 |
| Output $ / M tokens | $7.50 | $6 |
| Results tracked | 36 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
Mistral Medium: 34.2 (#243), Mixtral 8x22B: 24.2 (#329)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| WeirdML | 43.7% | 3.2% |
| LMArena Coding | 1434 | 1166 |
| FrontierCode | 8% | — |
| SciCode | 40.2% | — |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| ALE-Bench | 763.98 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Mistral Medium leads
Mistral Medium: 28.3 (#90), Mixtral 8x22B: 23.1 (#127)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.7% | — |
| Cybench | — | 7.5% |
Reasoning Mistral Medium leads
Mistral Medium: 24.0 (#167), Mixtral 8x22B: 19.9 (#248)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1150 |
| DTBench | 75.5% | 55.1% |
| Kagi LLM Benchmark | 50% | — |
| CritPt | 0% | — |
| LMCA | 26.1% | — |
| Surface Evolver Bench | 26.9% | — |
| Epoch Capabilities Index | — | 122.03 |
| ForecastBench | — | 56.3 |
Math Mistral Medium leads
Mistral Medium: 28.1 (#245), Mixtral 8x22B: 22.9 (#275)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1408 | 1184 |
| MATH Level 5 | 81.6% | 24.2% |
| OTIS Mock AIME 2024-2025 | 32.2% | — |
| ProofBench | 9% | — |
| Omni-MATH | — | 16.3% |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Mistral Medium leads
Mistral Medium: 25.0 (#265), Mixtral 8x22B: 15.1 (#293)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 59.5% | 34.1% |
| LMArena Expert | 1408 | 1113 |
| Humanity's Last Exam | 4.5% | — |
| MMLU-Pro | — | 46% |
| Vectara Hallucination Rate | 22.7% | — |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multimodal Not comparable
Mistral Medium: 35.3 (#88), Mixtral 8x22B: —
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Vision | 1172 | — |
Multilingual Mistral Medium leads
Mistral Medium: 52.1 (#91), Mixtral 8x22B: 32.8 (#255)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1408 | 1128 |
| LMArena Chinese | 1447 | 1116 |
| LMArena French | 1459 | 1166 |
| LMArena German | 1432 | 1141 |
| LMArena Japanese | 1378 | 1037 |
| LMArena Korean | 1380 | 1057 |
| LMArena Russian | 1411 | 1158 |
| LMArena Spanish | 1433 | 1151 |
Instruction Following Mistral Medium leads
Mistral Medium: 73.7 (#116), Mixtral 8x22B: 57.7 (#266)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1398 | 1147 |
| IFEval | — | 72.4% |
Long Context Mistral Medium leads
Mistral Medium: 42.9 (#114), Mixtral 8x22B: 34.7 (#247)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1406 | 1144 |
Writing & Preference Mistral Medium leads
Mistral Medium: 60.0 (#103), Mixtral 8x22B: 36.9 (#262)
| Benchmark | Mistral Medium | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1424 | 1162 |
| LMArena Creative Writing | 1391 | 1141 |
| LMArena Multi-Turn | 1418 | 1130 |
| Short-Story Creative Writing | 77.3% | — |
| WildBench | — | 71.1% |
Frequently asked questions
Is Mistral Medium better than Mixtral 8x22B?
Mistral Medium is the stronger model overall, scoring 36.3 to 27.1 on the Noometry Index.
Which is cheaper, Mistral Medium or Mixtral 8x22B?
Mixtral 8x22B is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Mixtral 8x22B better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 64K.
How many benchmarks do Mistral Medium and Mixtral 8x22B share?
21 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Mixtral 8x22B has 34.