Compare / DeepSeek V4.1 Flash vs Mistral Large 4
DeepSeek V4.1 Flash vs Mistral Large 4
37 results from 2 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | DeepSeek V4.1 Flash | Mistral Large 4 |
|---|---|---|---|
| CatalogBenchmeasured by us | Channel rules broken% of products | 1.8% | 1.2% |
| CatalogBenchmeasured by us | Claims to check% of products | 0.0% | 3.3% |
| CatalogBenchmeasured by us | Content quality% of checks | 96.2% | 95.9% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.6% | 0.0% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 3.6% | 3.0% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 47.0% | 10.7% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 23.2% | 1.8% |
| CatalogBenchmeasured by us | Unsupported claims% of products | 39.9% | 80.4% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 1.25 | 3.69 |
| CatalogBenchmeasured by us | Wrong attributes% of products | 20.2% | 17.3% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 2.4% | 1.8% |
| CatalogBenchmeasured by us | Channel compliance% of products | 99.4% | 100.0% |
| CatalogBenchmeasured by us | Channel rules broken% of products | 3.0% | 3.0% |
| CatalogBenchmeasured by us | Claims to check% of products | 0.0% | 0.7% |
| CatalogBenchmeasured by us | Conflicts caught% of conflicts | 100.0% | 100.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 96.7% | 96.5% |
| CatalogBenchmeasured by us | Cost per productUS dollars | $0.0061 | $0.0033 |
| CatalogBenchmeasured by us | Decision accuracy% of decisions | 96.9% | 93.1% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Field accuracy% of missing fields | 94.4% | 94.4% |
| CatalogBenchmeasured by us | Invented values% of filled values | 3.1% | 3.5% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 3.6% | 4.2% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 63.7% | 39.9% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 41.1% | 10.7% |
| CatalogBenchmeasured by us | Unsupported claims% of products | 13.1% | 33.3% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 0.54 | 0.91 |
| CatalogBenchmeasured by us | Wrong attributes% of products | 20.2% | 19.6% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 3.0% | 1.2% |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 39.5 | 38.4 |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 39.2% | 35.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 84.0% | 81.3% |
| Artificial Analysisreported | Output speedtokens per second | 213 | 106 |
| Artificial Analysisreported | SciCode% of problems | 51.9% | 54.2% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 26.8% | 26.8% |
| Artificial Analysisreported | Time to first answer tokenseconds | 0.89 | 19.8 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.