Model comparison
DeepSeek-V3.1 vs Ministral 8B
DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 27.8 on the Noometry Index. Ministral 8B costs 4.3× less per token, which makes it the better buy when DeepSeek-V3.1's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 14 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Ministral 8B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.0 to 12.5.
- The biggest single-benchmark swing is DTBench: 82.7% for DeepSeek-V3.1 and 45.7% for Ministral 8B.
- Ministral 8B is cheaper at $0.10 / $0.10 per million input/output tokens, against $0.25 / $0.95 for DeepSeek-V3.1.
- DeepSeek-V3.1 accepts more context: 164K tokens versus 128K.
Side by side
| DeepSeek-V3.1 | Ministral 8B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 42.2 | 27.8 |
| Released | 2025-08-21 | 2024-10-01 |
| Weights | Open | Open |
| Context window | 164K | 128K |
| Max output | 8K | 128K |
| Input $ / M tokens | $0.25 | $0.10 |
| Output $ / M tokens | $0.95 | $0.10 |
| Results tracked | 27 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 39.8 (#140), Ministral 8B: 34.4 (#226)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Coding | 1417 | 1202 |
| WeirdML | 38.4% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, Ministral 8B: 16.2 (#140)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.3 (#109), Ministral 8B: 17.7 (#279)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1191 |
| DTBench | 82.7% | 45.7% |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.1 (#125), Ministral 8B: 23.7 (#262)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Math | 1420 | 1188 |
| MATH Level 5 | — | 14.9% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.0 (#86), Ministral 8B: 12.5 (#292)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 7.4% |
| LMArena Expert | 1403 | 1170 |
| GPQA Diamond | — | 27.1% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.8 (#100), Ministral 8B: 35.3 (#242)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1400 | 1165 |
| LMArena Chinese | 1470 | 1193 |
| LMArena Russian | 1405 | 1195 |
| LMArena French | 1447 | — |
| LMArena German | 1412 | — |
| LMArena Japanese | 1374 | — |
| LMArena Korean | 1337 | — |
| LMArena Spanish | 1429 | — |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 74.1 (#105), Ministral 8B: 60.7 (#245)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1161 |
Long Context Too close to call
DeepSeek-V3.1: 36.3 (#227), Ministral 8B: 36.7 (#222)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1422 | 1212 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.5 (#93), Ministral 8B: 39.8 (#241)
| Benchmark | DeepSeek-V3.1 | Ministral 8B |
|---|---|---|
| LMArena Text | 1420 | 1191 |
| LMArena Creative Writing | 1401 | 1175 |
| LMArena Multi-Turn | 1408 | 1166 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Ministral 8B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 27.8 on the Noometry Index. Ministral 8B costs 4.3× less per token, which makes it the better buy when DeepSeek-V3.1's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1 or Ministral 8B?
Ministral 8B is cheaper. It lists at $0.10 per million input tokens and $0.10 per million output tokens; DeepSeek-V3.1 lists at $0.25 and $0.95.
Is DeepSeek-V3.1 or Ministral 8B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 39.8 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1 does, with 164K tokens against 128K.
How many benchmarks do DeepSeek-V3.1 and Ministral 8B share?
14 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Ministral 8B has 17.