Model comparison
DeepSeek-V3.1-Terminus vs Mixtral 8x22B
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 26.7 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 11 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Mixtral 8x22B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 37.1.
- The biggest single-benchmark swing is DTBench: 81.3% for DeepSeek-V3.1-Terminus and 55.1% for Mixtral 8x22B.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- DeepSeek-V3.1-Terminus accepts more context: 164K tokens versus 64K.
Side by side
| DeepSeek-V3.1-Terminus | Mixtral 8x22B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 42.6 | 26.7 |
| Released | 2025-09-22 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 164K | 64K |
| Max output | 66K | 64K |
| Input $ / M tokens | $0.30 | $2 |
| Output $ / M tokens | $1 | $6 |
| Results tracked | 16 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 41.5 (#114), Mixtral 8x22B: 23.8 (#324)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1426 | 1166 |
| SciCode | 40.6% | — |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| ALE-Bench | 745.17 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Mixtral 8x22B: 22.8 (#119)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| Cybench | — | 7.5% |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 25.7 (#126), Mixtral 8x22B: 19.1 (#246)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1427 | 1150 |
| DTBench | 81.3% | 55.1% |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 122 |
| ForecastBench | — | 56.3 |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 35.7 (#138), Mixtral 8x22B: 21.2 (#271)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1402 | 1184 |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Mixtral 8x22B: 15.1 (#288)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | — | 34.1% |
| MMLU-Pro | — | 46% |
| GPQA (HELM) | — | 33.4% |
| LMArena Expert | — | 1113 |
| MMLU | — | 77.8% |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.3 (#87), Mixtral 8x22B: 32.9 (#250)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1407 | 1128 |
| LMArena Russian | 1434 | 1158 |
| LMArena Chinese | — | 1116 |
| LMArena French | — | 1165 |
| LMArena German | — | 1141 |
| LMArena Japanese | — | 1037 |
| LMArena Korean | — | 1056 |
| LMArena Spanish | — | 1150 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.4 (#96), Mixtral 8x22B: 57.9 (#261)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1406 | 1147 |
| IFEval | — | 72.4% |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.5 (#88), Mixtral 8x22B: 34.6 (#242)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1426 | 1144 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.3 (#87), Mixtral 8x22B: 37.1 (#257)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1420 | 1162 |
| LMArena Creative Writing | 1404 | 1141 |
| LMArena Multi-Turn | 1411 | 1130 |
| WildBench | — | 71.1% |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Mixtral 8x22B?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 26.7 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1-Terminus or Mixtral 8x22B?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is DeepSeek-V3.1-Terminus or Mixtral 8x22B better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 23.8 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1-Terminus does, with 164K tokens against 64K.
How many benchmarks do DeepSeek-V3.1-Terminus and Mixtral 8x22B share?
11 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Mixtral 8x22B has 34.