Compare / Gemini 3.8 Flash vs Mistral Large 4
Gemini 3.8 Flash vs Mistral Large 4
Which is cheaper, which is more capable, and which does better at each kind of work, from 94 results on 6 sources that measured both models.
Of the 94 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331). See also Google vs Mistral AI, every model from both labs.
Gemini 3.8 Flash or Mistral Large 4? The short answer
Pick Gemini 3.8 Flash if quality matters most: it scores higher (Spring Prompt overall 57 against 24). Pick Mistral Large 4 if cost matters more: it costs 3.1 times less per product listing on CatalogBench. Gemini 3.8 Flash is also faster.
- Spring Prompt overall: Gemini 3.8 Flash 57, Mistral Large 4 24 (out of 100; how the score works).
- Mistral Large 4 is cheaper to run: $0.003 against $0.010 per product listing on CatalogBench (3.1 times less).
- Gemini 3.8 Flash scores higher on the Artificial Analysis Intelligence Index: 40.9 against 38.4.
- Gemini 3.8 Flash writes faster: 141 tokens a second against 60.
- Mistral Large 4 is newer: released 6 Oct 2026, against 2 Sep 2026 for Gemini 3.8 Flash.
- On our own benchmarks, on the results both models have: Gemini 3.8 Flash does better for product listings, decks from an analysis and marketing planning.
- On third-party benchmarks, on the results both models have: Gemini 3.8 Flash does better for reasoning and knowledge; Mistral Large 4 for agents and tool use.
Head to head
Every fact side by side
| Gemini 3.8 Flash | Mistral Large 4 | |
|---|---|---|
| Price per million tokens | $1.50 in · $7.50 out | $1.36 in · $4.18 out |
| Blended (3 in : 1 out) | $3.00 | $2.06 |
| Cheapest host, blended | $1.50 (Google) | $1.05 (Mistral) |
| Cost per product listing (CatalogBench) | $0.010 | $0.003 |
| Intelligence Index | 40.9 | 38.4 |
| Spring Prompt overall | 57 | 24 |
| Output speed, tokens a second | 141 | 60 |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Released | 2 Sep 2026 | 6 Oct 2026 (newer) |
| Weights | Closed | Open |
| Developer based in | the US | France |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Their slides, side by side
The same brief for both: Manchester store investment review for Hearthside Coffee, an invented company. First slide of each deck from DeckBench.
From our research
- Gemini 4 Hydrogen got faster: we reran our speed chess test 16 Nov 2026
- Muse Ember review: GPT-6 Luna's price, better at analysis, and no Gemini Flash against a clock 10 Nov 2026
- Kimi K3.1 Pro first look: the best open-weight model on our board, and very slow 5 Nov 2026
- DeepSeek-V4.1-Flash-1017 review: better at analysis, much worse at sticking to the product data 19 Oct 2026
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Gemini 3.8 Flash | Mistral Large 4 |
|---|---|---|
| BulletBench measured by Spring Prompt · 3 results | ||
| Cost per game · Blitz 3+2US dollars, lower is better 11 · 2 of 23 measured | $0.0795 | $0.0082 |
| Ladder Elo · Lightning 10+1ladder Elo, higher is better 40 · 39 of 45 measured | 197low reasoning | 249 |
| Cost per game · Lightning 10+1US dollars, lower is better 30 · 40 of 45 measured | $0.0041low reasoning | $0.0117 |
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 19 · 8 of 36 measured | $0.0103 | $0.0033 |
| Reliably publish-ready% of products, higher is better 23 · 31 of 35 measured | 35.7% | 10.7%high reasoning |
| Reliably publish-ready · sales brief% of products, higher is better 26 · 28 of 35 measured | 3.6% | 1.8%high reasoning |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 15 · 8 of 36 measured | $0.0613 | $0.0148 |
| Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 29 · 34 of 36 measured | 771 | 595 |
| Accurate decks% of tasks, higher is better 27 · 31 of 36 measured | 16.7% | 0.0% |
| ROASBench measured by Spring Prompt · 3 results | ||
| Overall scorescore out of 100, higher is better 13 · 26 of 36 measured | 51.4 | 37.0 |
| Audience scorescore out of 100, higher is better 13 · 25 of 36 measured | 67.1 | 55.8 |
| Business scorescore out of 100, higher is better 12 · 26 of 36 measured | 51.0 | 34.3 |
| SurveyBench measured by Spring Prompt · 3 results | ||
| Analysis ratingrating, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 18 · 19 of 27 measured | 845 | 770 |
| Cost per taskUS dollars, lower is better 14 · 13 of 27 measured | $0.0943 | $0.0924 |
| Analysis scorepoints of 100, higher is better 17 · 19 of 27 measured | 74.7 | 63.8 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 25 · 36 of 249 measured | 40.9high reasoning | 38.4 |
| Humanity's Last Exam% of questions, higher is better 18 · 65 of 250 measured | 47.8%high reasoning | 35.0% |
| Terminal-Bench 4.0% of tasks, higher is better 30 · 26 of 120 measured | 19.7%high reasoning | 24.7% |
Show all 94 results (76 more)
| Measure | Gemini 3.8 Flash | Mistral Large 4 |
|---|---|---|
| BulletBench measured by Spring Prompt · 7 results | ||
| Games lost on time · Blitz 3+2% of games, lower is better 9 · 1 of 23 measured | 12.5% | 0.0% |
| Games lost on time · Lightning 10+1% of games, lower is better 42 · 36 of 45 measured | 83.3%low reasoning | 58.3% |
| Invalid moves · Blitz 3+2% of moves, lower is better 1 · 22 of 23 measured | 0.0% | 0.6% |
| Invalid moves · Lightning 10+1% of moves, lower is better 1 · 43 of 45 measured | 0.0%low reasoning | 1.8% |
| Ladder Elo · Blitz 3+2ladder Elo, higher is better 2 · 18 of 23 measured | 1,126 | 555 |
| Median move time · Blitz 3+2milliseconds, lower is better 10 · 3 of 23 measured | 4.0 s | 0.8 s |
| Median move time · Lightning 10+1milliseconds, lower is better 44 · 30 of 45 measured | 1.8 slow reasoning | 0.8 s |
| CatalogBench measured by Spring Prompt · 28 results | ||
| Channel compliance% of products, higher is better 28 · 1 of 36 measured | 99.4% | 100.0% |
| Channel rules broken% of products, lower is better 22 · 35 of 36 measured | 0.6% | 3.0% |
| Channel rules broken · sales brief% of products, lower is better 19 · 28 of 36 measured | 0.6% | 1.2% |
| Claims to check% of products, lower is better 1 · 27 of 34 measured | 0.0% | 0.7%high reasoning |
| Claims to check · sales brief% of products, lower is better 29 · 27 of 34 measured | 5.4% | 3.3%high reasoning |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 36 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 11 · 23 of 36 measured | 97.7% | 96.5% |
| Content quality · sales brief% of checks, higher is better 17 · 28 of 36 measured | 97.6% | 95.9% |
| Decision accuracy% of decisions, higher is better 21 · 33 of 36 measured | 96.4% | 93.1% |
| Failed outputs% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Failed outputs · sales brief% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Field accuracy% of missing fields, higher is better 27 · 10 of 36 measured | 92.7% | 94.4% |
| Invented values% of filled values, lower is better 4 · 32 of 36 measured | 0.3% | 3.5% |
| Missing UK information% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 30 · 18 of 36 measured | 8.3% | 4.2% |
| Not findable · sales brief% of products, lower is better 31 · 15 of 36 measured | 8.9% | 3.0%high reasoning |
| Publish-ready listings% of products, higher is better 24 · 30 of 36 measured | 54.8% | 39.9%high reasoning |
| Publish-ready listings · sales brief% of products, higher is better 26 · 29 of 36 measured | 16.7% | 10.7%high reasoning |
| Time per productseconds, lower is better 21 · 8 of 36 measured | 18.7 | 8.20 |
| Unsupported claimsclaims per product, lower is better 25 · 26 of 36 measured | 0.79 | 0.91high reasoning |
| Unsupported claims% of products, lower is better 27 · 29 of 36 measured | 31.6% | 33.3%high reasoning |
| Unsupported claims · sales briefclaims per product, lower is better 27 · 31 of 36 measured | 3.08 | 3.69high reasoning |
| Unsupported claims · sales brief% of products, lower is better 31 · 30 of 36 measured | 81.6% | 80.4%high reasoning |
| Wrong attributes% of products, lower is better 14 · 26 of 36 measured | 15.5% | 19.6%high reasoning |
| Wrong attributes · sales brief% of products, lower is better 2 · 20 of 36 measured | 11.9% | 17.3%high reasoning |
| Wrong category or variant% of products, lower is better 30 · 5 of 36 measured | 3.6% | 1.2% |
| Wrong category or variant · sales brief% of products, lower is better 26 · 7 of 36 measured | 3.0% | 1.8% |
| DeckBench measured by Spring Prompt · 14 results | ||
| Caveat dropped% of tasks, lower is better 1 · 16 of 36 measured | 0.0% | 16.7% |
| Clean layout% of tasks, higher is better 26 · 18 of 36 measured | 16.7% | 33.3% |
| Design quality% of the maximum, higher is better 21 · 35 of 36 measured | 61.7% | 30.5% |
| Draft figure quoted% of tasks, lower is better 29 · 15 of 36 measured | 50.0% | 16.7% |
| Findings missing% of tasks, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 29 · 35 of 36 measured | 37.8% | 24.7% |
| Layout defects% of tasks, lower is better 26 · 18 of 36 measured | 83.3% | 66.7% |
| Misleading metric used% of tasks, lower is better 1 · 1 of 36 measured | 0.0% | 0.0% |
| Presentable decks% of tasks, higher is better 17 · 17 of 36 measured | 0.0% | 0.0% |
| Recommendation late or wrong% of tasks, lower is better 22 · 33 of 36 measured | 33.3% | 83.3% |
| Slides needing work% of tasks, lower is better 12 · 36 of 36 measured | 33.3% | 100.0% |
| Time per deckseconds, lower is better 15 · 18 of 36 measured | 93.0 | 114 |
| Unsupported claims% of tasks, lower is better 32 · 32 of 36 measured | 50.0% | 50.0% |
| Unsupported numbers% of tasks, lower is better 28 · 34 of 36 measured | 16.7% | 33.3% |
| ROASBench measured by Spring Prompt · 6 results | ||
| Consistency scorescore out of 100, higher is better 8 · 30 of 36 measured | 36.4 | 16.7 |
| Contribution profitsimulated US dollars, higher is better 5 · 23 of 36 measured | $842,432 | $533,401 |
| Cost of a runUS dollars, lower is better 15 · 9 of 36 measured | $0.16 | $0.0657 |
| Months over budgetmonths of 12, lower is better 1 · 1 of 36 measured | 0 | 0high reasoning |
| Planning scorescore out of 100, higher is better 23 · 7 of 36 measured | 54.7 | 55.0high reasoning |
| Return on ad spendprofit per $1 spent, higher is better 13 · 23 of 36 measured | $0.65 | $0.45 |
| SurveyBench measured by Spring Prompt · 17 results | ||
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 7 · 21 of 27 measured | 60.9% | 35.5% |
| Leading or double-barrelled questions% of tasks, lower is better 24 · 12 of 27 measured | 66.7% | 33.3% |
| Mistaken next steps% of tasks, lower is better 5 · 23 of 27 measured | 11.1% | 66.7% |
| Numbers right% of answers, higher is better 1 · 19 of 27 measured | 100.0% | 98.9% |
| Objectives unanswerable% of tasks, lower is better 1 · 18 of 27 measured | 0.0% | 11.1% |
| Precision ignored% of tasks, lower is better 21 · 23 of 27 measured | 44.4% | 55.6% |
| Research scorepoints of 100, higher is better 17 · 19 of 27 measured | 77.6 | 71.2 |
| Sound research% of tasks, higher is better 17 · 17 of 27 measured | 0.0% | 0.0% |
| Stakeholder's view presumed% of tasks, lower is better 18 · 18 of 27 measured | 22.2% | 22.2% |
| Survey plan scorepoints of 100, higher is better 21 · 19 of 27 measured | 84.4 | 88.4 |
| Survey structure% of tasks, lower is better 12 · 3 of 27 measured | 22.2% | 11.1% |
| Time per taskseconds, lower is better 10 · 25 of 27 measured | 131 | 439 |
| Traps handled% of traps, higher is better 21 · 20 of 27 measured | 78.3% | 78.7% |
| Traps missed% of tasks, lower is better 19 · 25 of 27 measured | 88.9% | 100.0% |
| Unsupported findings% of tasks, lower is better 16 · 24 of 27 measured | 77.8% | 100.0% |
| Wrong numbers% of tasks, lower is better 1 · 17 of 27 measured | 0.0% | 22.2% |
| Wrong recommendation% of tasks, lower is better 1 · 19 of 27 measured | 0.0% | 44.4% |
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| Long-context reasoning (AA-LCR)% of questions, higher is better 7 · 32 of 241 measured | 84.0%medium reasoning | 81.3% |
| Output speedtokens per second, higher is better 31 · 61 of 78 measured | 141high reasoning | 60.2 |
| SciCode% of problems, higher is better 20 · 34 of 121 measured | 56.6%high reasoning | 54.2% |
| Time to first answer tokenseconds, lower is better 58 · 68 of 78 measured | 15.5high reasoning | 33.9 |
Compare them with other models
- Google vs Mistral AIEvery model from both labs, best against best
- Gemini 3.8 Flash vs Claude Opus 5.5Most intelligent at $3–10 per million tokens
- Gemini 3.8 Flash vs DeepSeek-V4.1-FlashFastest with similar intelligence
- Gemini 3.8 Flash vs GPT-6.2 SolHighest-scoring overall at $3–10 per million tokens
- Mistral Large 4 vs Claude Opus 5.5Most intelligent available today
Quick answers
Is Gemini 3.8 Flash better than Mistral Large 4?
Pick Gemini 3.8 Flash if quality matters most: it scores higher (Spring Prompt overall 57 against 24). Pick Mistral Large 4 if cost matters more: it costs 3.1 times less per product listing on CatalogBench. Gemini 3.8 Flash is also faster.
Which is cheaper, Gemini 3.8 Flash or Mistral Large 4?
Mistral Large 4 is cheaper to run: $0.003 against $0.010 per product listing on CatalogBench (3.1 times less).
Which is faster, Gemini 3.8 Flash or Mistral Large 4?
Gemini 3.8 Flash writes about 141 tokens a second on its usual API, against 60 for Mistral Large 4 (Artificial Analysis).
Which is better for product listings, Gemini 3.8 Flash or Mistral Large 4?
Gemini 3.8 Flash ranks 25 of 35 models on sale for turning a sparse product feed and photos into listings that can go live, against 30 for Mistral Large 4.
Which is better for decks from an analysis, Gemini 3.8 Flash or Mistral Large 4?
Gemini 3.8 Flash ranks 29 of 35 models on sale for turning a finished analysis into a deck you could present as it is, against 33 for Mistral Large 4.
Which is better for user surveys, Gemini 3.8 Flash or Mistral Large 4?
Gemini 3.8 Flash ranks 18 of 26 models on sale for planning a user survey and reading its results without being misled, against 19 for Mistral Large 4.
Which is better for marketing planning, Gemini 3.8 Flash or Mistral Large 4?
Gemini 3.8 Flash ranks 13 of 35 models on sale for planning a year of ad spend without overspending, against 26 for Mistral Large 4.
Run Gemini 3.8 Flash and Mistral Large 4 on your own prompt.
Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.

