Model comparison
DeepSeek V4 Flash vs Mistral Large 3
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 38.6 on the Noometry Index.
Last verified . 21 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 21 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 8 categories and Mistral Large 3 in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 14.6.
- The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4 Flash and 7.5% for Mistral Large 3.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.25 / $0.75 for Mistral Large 3.
- DeepSeek V4 Flash accepts more context: 1M tokens versus 262K.
Side by side
| DeepSeek V4 Flash | Mistral Large 3 | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 53.6 | 38.6 |
| Released | 2026-04-24 | 2025-12-02 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 393K | 8K |
| Input $ / M tokens | $0.15 | $0.25 |
| Output $ / M tokens | $0.60 | $0.75 |
| Results tracked | 41 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
DeepSeek V4 Flash: 47.6 (#56), Mistral Large 3: 34.1 (#230)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena WebDev | 1581 | 1230 |
| LMArena Coding | 1457 | 1447 |
| FrontierCode | 18.8% | — |
| SciCode | 49.9% | — |
| WeirdML | 63% | — |
| ALE-Bench | 1,306 | — |
Reasoning DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.7 (#25), Mistral Large 3: 14.6 (#314)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| Kagi LLM Benchmark | 52.2% | 50.9% |
| NYT Connections (extended) | 89.6% | 7.5% |
| LMArena Hard Prompts | 1443 | 1429 |
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| ARC-AGI-1 | 89% | — |
| CritPt | 16.6% | — |
| Chess Puzzles | 33% | — |
| Thematic Generalization | — | 23% |
| Mystery Game Puzzles | 34% | — |
| DTBench | 90.9% | — |
| LMCA | 41.7% | — |
| Epoch Capabilities Index | 154.51 | — |
Math DeepSeek V4 Flash leads
DeepSeek V4 Flash: 60.0 (#31), Mistral Large 3: 36.0 (#131)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Math | 1427 | 1415 |
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| OTIS Mock AIME 2024-2025 | 94.4% | — |
| ProofBench | 56% | — |
Knowledge DeepSeek V4 Flash leads
DeepSeek V4 Flash: 55.5 (#43), Mistral Large 3: 35.5 (#173)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Expert | 1441 | 1421 |
| GPQA Diamond | 91% | — |
| SimpleQA Verified | 33.6% | — |
| Vectara Hallucination Rate | — | 14.5% |
Multimodal Not comparable
DeepSeek V4 Flash: —, Mistral Large 3: 37.6 (#65)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Vision | — | 1222 |
Multilingual Too close to call
DeepSeek V4 Flash: 53.2 (#71), Mistral Large 3: 52.6 (#79)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Non-English | 1420 | 1412 |
| LMArena Chinese | 1467 | 1448 |
| LMArena French | 1438 | 1456 |
| LMArena German | 1418 | 1438 |
| LMArena Japanese | 1403 | 1399 |
| LMArena Korean | 1383 | 1384 |
| LMArena Russian | 1428 | 1412 |
| LMArena Spanish | 1437 | 1437 |
Instruction Following Too close to call
DeepSeek V4 Flash: 75.1 (#76), Mistral Large 3: 74.1 (#104)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Instruction Following | 1421 | 1402 |
Long Context Too close to call
DeepSeek V4 Flash: 43.8 (#81), Mistral Large 3: 43.1 (#102)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Longer Query | 1435 | 1413 |
Writing & Preference DeepSeek V4 Flash leads
DeepSeek V4 Flash: 64.0 (#59), Mistral Large 3: 60.2 (#100)
| Benchmark | DeepSeek V4 Flash | Mistral Large 3 |
|---|---|---|
| LMArena Text | 1432 | 1427 |
| LMArena Creative Writing | 1403 | 1385 |
| EQ-Bench Creative Writing | 1559 | 1412 |
| LMArena Multi-Turn | 1449 | 1429 |
Frequently asked questions
Is DeepSeek V4 Flash better than Mistral Large 3?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 38.6 on the Noometry Index.
Which is cheaper, DeepSeek V4 Flash or Mistral Large 3?
DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Mistral Large 3 lists at $0.25 and $0.75.
Is DeepSeek V4 Flash or Mistral Large 3 better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.6 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Flash does, with 1M tokens against 262K.
How many benchmarks do DeepSeek V4 Flash and Mistral Large 3 share?
21 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Mistral Large 3 has 24.