Compare / Mistral Large 4 vs GPT-6.1 Sol
Mistral Large 4 vs GPT-6.1 Sol
53 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Mistral Large 4 | GPT-6.1 Sol |
|---|---|---|---|
| CatalogBenchmeasured by us | Channel rules broken% of products | 1.2% | 0.0% |
| CatalogBenchmeasured by us | Claims to check% of products | 3.3% | 0.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 95.9% | 98.8% |
| CatalogBenchmeasured by us | Failed outputs% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 3.0% | 5.4% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 10.7% | 72.0% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 1.8% | 67.9% |
| CatalogBenchmeasured by us | Unsupported claims% of products | 80.4% | 3.6% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 3.69 | 0.07 |
| CatalogBenchmeasured by us | Wrong attributes% of products | 17.3% | 19.1% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 1.8% | 1.8% |
| CatalogBenchmeasured by us | Channel compliance% of products | 100.0% | 100.0% |
| CatalogBenchmeasured by us | Channel rules broken% of products | 3.0% | 0.0% |
| CatalogBenchmeasured by us | Claims to check% of products | 0.7% | 0.0% |
| CatalogBenchmeasured by us | Conflicts caught% of conflicts | 100.0% | 100.0% |
| CatalogBenchmeasured by us | Content quality% of checks | 96.5% | 98.6% |
| CatalogBenchmeasured by us | Cost per productUS dollars | $0.0033 | $0.0084 |
| CatalogBenchmeasured by us | Decision accuracy% of decisions | 93.1% | 98.7% |
| 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 | 3.5% | 2.2% |
| CatalogBenchmeasured by us | Missing UK information% of products | 0.0% | 0.0% |
| CatalogBenchmeasured by us | Not findable% of products | 4.2% | 4.8% |
| CatalogBenchmeasured by us | Publish-ready listings% of products | 39.9% | 78.0% |
| CatalogBenchmeasured by us | Reliably publish-ready% of products | 10.7% | 75.0% |
| CatalogBenchmeasured by us | Unsupported claimsclaims per product | 0.91 | 0.00 |
| CatalogBenchmeasured by us | Unsupported claims% of products | 33.3% | 0.0% |
| CatalogBenchmeasured by us | Wrong attributes% of products | 19.6% | 16.7% |
| CatalogBenchmeasured by us | Wrong category or variant% of products | 1.2% | 1.8% |
| DeckBenchmeasured by us | Accurate decks% of tasks | 0.0% | 100.0% |
| DeckBenchmeasured by us | Caveat dropped% of tasks | 16.7% | 0.0% |
| DeckBenchmeasured by us | Clean layout% of tasks | 33.3% | 83.3% |
| DeckBenchmeasured by us | Cost per deckUS dollars | $0.0148 | $0.15 |
| DeckBenchmeasured by us | Deck ratingrating | 515 | 1,412 |
| DeckBenchmeasured by us | Design quality% of the maximum | 30.5% | 71.4% |
| DeckBenchmeasured by us | Draft figure quoted% of tasks | 16.7% | 0.0% |
| DeckBenchmeasured by us | Findings missing% of tasks | 0.0% | 0.0% |
| DeckBenchmeasured by us | Head-to-head win rate% of comparisons | 10.7% | 74.6% |
| DeckBenchmeasured by us | Layout defects% of tasks | 66.7% | 16.7% |
| DeckBenchmeasured by us | Misleading metric used% of tasks | 0.0% | 0.0% |
| DeckBenchmeasured by us | Presentable decks% of tasks | 0.0% | 50.0% |
| DeckBenchmeasured by us | Recommendation late or wrong% of tasks | 83.3% | 0.0% |
| DeckBenchmeasured by us | Slides needing work% of tasks | 100.0% | 50.0% |
| DeckBenchmeasured by us | Unsupported claims% of tasks | 50.0% | 0.0% |
| DeckBenchmeasured by us | Unsupported numbers% of tasks | 33.3% | 0.0% |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 38.4 | 51.8 |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 35.0% | 52.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 81.3% | 84.0% |
| Artificial Analysisreported | Output speedtokens per second | 106 | 57.2 |
| Artificial Analysisreported | SciCode% of problems | 54.2% | 55.8% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 26.8% | 56.1% |
| Artificial Analysisreported | Time to first answer tokenseconds | 19.8 | 1.81 |
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.