Model comparison
DeepSeek-V3.1-Terminus vs Mixtral 8x7B
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 27.1 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 8x7B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.0 to 34.2.
- The biggest single-benchmark swing is DTBench: 81.3% for DeepSeek-V3.1-Terminus and 49.6% for Mixtral 8x7B.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $0.70 / $0.70 for Mixtral 8x7B.
- DeepSeek-V3.1-Terminus accepts more context: 164K tokens versus 32K.
Side by side
| DeepSeek-V3.1-Terminus | Mixtral 8x7B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 43.1 | 27.1 |
| Released | 2025-09-22 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 164K | 32K |
| Max output | 147K | 32K |
| Input $ / M tokens | $0.27 | $0.70 |
| Output $ / M tokens | $1 | $0.70 |
| Results tracked | 16 | 38 |
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: 42.0 (#113), Mixtral 8x7B: 32.8 (#269)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | 1426 | 1126 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
| HumanEval+ | — | 39.6% |
| MBPP+ | — | 49.7% |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Mixtral 8x7B: 18.2 (#285)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1115 |
| DTBench | 81.3% | 49.6% |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| LMCA | 28.6% | — |
| Adversarial NLI | — | 55.2% |
| Epoch Capabilities Index | — | 118.47 |
| ForecastBench | — | 56.3 |
| HellaSwag | — | 86.7% |
| PIQA | — | 83.6% |
| WinoGrande | — | 77.2% |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Mixtral 8x7B: 18.8 (#289)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Math | 1402 | 1147 |
| Omni-MATH | — | 10.5% |
| MATH Level 5 | — | 10% |
| GSM8K | — | 74.4% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Mixtral 8x7B: 11.0 (#301)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| GPQA Diamond | — | 30.6% |
| MMLU-Pro | — | 33.5% |
| GPQA (HELM) | — | 29.6% |
| LMArena Expert | — | 1088 |
| ARC (AI2) Challenge | — | 87.3% |
| MMLU | — | 70.6% |
| OpenBookQA | — | 85.8% |
| TriviaQA | — | 82.2% |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Mixtral 8x7B: 29.6 (#266)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 1407 | 1077 |
| LMArena Russian | 1436 | 1090 |
| LMArena Chinese | — | 1055 |
| LMArena French | — | 1166 |
| LMArena German | — | 1114 |
| LMArena Japanese | — | 931 |
| LMArena Korean | — | 968 |
| LMArena Spanish | — | 1111 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Mixtral 8x7B: 51.0 (#297)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Instruction Following | 1404 | 1109 |
| IFEval | — | 57.5% |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Mixtral 8x7B: 33.4 (#260)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | 1421 | 1103 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Mixtral 8x7B: 34.2 (#270)
| Benchmark | DeepSeek-V3.1-Terminus | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1419 | 1132 |
| LMArena Creative Writing | 1403 | 1109 |
| LMArena Multi-Turn | 1411 | 1115 |
| WildBench | — | 67.3% |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Mixtral 8x7B?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 27.1 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1-Terminus or Mixtral 8x7B?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Mixtral 8x7B lists at $0.70 and $0.70.
Is DeepSeek-V3.1-Terminus or Mixtral 8x7B better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 32.8 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1-Terminus does, with 164K tokens against 32K.
How many benchmarks do DeepSeek-V3.1-Terminus and Mixtral 8x7B share?
11 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Mixtral 8x7B has 38.