Models / Mistral Large 4

Mistral AI

Mistral Large 4

Mistral Large 4 is best for agents and tool use; a strong pick at a modest price. The cheapest model of similar intelligence is MiMo-V2.6-Flash, at $0.18 per million tokens against $2.06.

Results 19 Nov 2026 · 135 results · 6 sources

Test Mistral Large 4 on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$1.36 in · $4.18 out
Mistral AI's list price, 10 Nov 2026 · compare prices50% off in batch
Speed
60 tokens a second; on its test prompt the answer starts after 33.9 s, thinking time included
Artificial Analysis, on its usual API
Intelligence Index
38.4
Artificial Analysis
Spring Prompt overall
35th of 39 score 24 of 100 · how it works
Spring Prompt Score
22 of 100 33rd of 35 · 5 of 5 benchmarks · our own tests only
Developer
Mistral AI, based in France
Weights published
Released
6 Oct 2026
1,048,576 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 34th of 185Claude Opus 5.5 35th of 38Claude Fable 5.2 $2.06 blended, 70th percentilecheapest: Mistral Nemo 51st of 65Nemotron 3.5 Super VL
Models priced $1–3 per million tokens 5th of 34Muse Spark 1.3 6th of 8GPT-6.1 Terra sets this group 6th of 9Muse Spark 1.3
Models with similar intelligence sets this group 5th of 5Gemini 3.8 Flash $2.06 blended, 50th percentilecheapest: MiMo-V2.6-Flash 10th of 12DeepSeek-V4.1-Flash
Models writing 50–100 tokens a second 8th of 21Claude Opus 5.5 7th of 7Claude Opus 5.5 $2.06 blended, 65th percentilecheapest: Granite 4.2 8B sets this group
Models from developers based in the EUall Mistral AI's 1st of 15leads – $2.06 blended, 80th percentilecheapest: Mistral Nemo 6th of 6Ministral 3 3B
Open-weights models 10th of 104MiMo-V2.6-Pro 8th of 8Kimi K3.1 Pro $2.06 blended, 94th percentilecheapest: Mistral Nemo 32nd of 43Nemotron 3.5 Super VL

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

Measured cost per task

API cost recorded in our benchmark runs, excluding judging.

TaskBenchmarkCost per task
Product listingCatalogBench≈ $0.003
DeckDeckBench≈ $0.015
Survey plan and analysisSurveyBench≈ $0.092
Game of blitz chessBulletBench≈ $0.008

Cheapest model that passes

On our benchmarks that define a pass for each task: what one passing answer cost in our run, failed attempts included, and whether a cheaper model passed.

  • CatalogBench · cost per 1,000 accepted listings
    Mistral Large 4: $17.76 (31 of 168 attempts), 12 of 27. Cheapest: GPT-6 Luna at $0.70 (114 of 168 attempts). It passed at least as often.
  • DeckBench · cost per presentable deck
    Mistral Large 4: no presentable decks (0 of 6 tasks). Cheapest: GPT-6.2 Sol at $0.09 (5 of 6 tasks).
  • SurveyBench · cost per sound survey task
    Mistral Large 4: no sound survey tasks (0 of 9 tasks). Cheapest: GPT-6 Luna at $0.03 (3 of 9 tasks).

Cost per attempt divided by the share that passed, at the run date's prices, excluding judging. Each model at its cheapest setting per passing answer. With a handful of tasks, a pass or two either way is within chance. No estimate where a run had missing outputs, because their cost may not have been recorded.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $1–3 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live 30th of 35 available · 4th of 6 at its price
BenchmarkMistral Large 4Rank among availableBest available
CatalogBenchCatalogBench: reliably publish-ready 10.7% 31st of 35 Gemini 4 Neon 78.6%
CatalogBenchCatalogBench (sales brief): reliably publish-ready 1.8% 28th of 35 GPT-6.2 Sol 75.0%
Decks from an analysisTurning a finished analysis into a deck you could present as it is 33rd of 35 available · 5th of 6 at its price
BenchmarkMistral Large 4Rank among availableBest available
DeckBenchDeckBench: deck rating 595 33rd of 35 Qwen4-Max 1,563
User surveysPlanning a user survey and reading its results without being misled 19th of 26 available · 3rd of 4 at its price
BenchmarkMistral Large 4Rank among availableBest available
SurveyBenchSurveyBench: analysis rating 770 19th of 26 Claude Fable 5.2 1,670
Marketing planningPlanning a year of ad spend without overspending 26th of 35 available · 3rd of 7 at its price
BenchmarkMistral Large 4Rank among availableBest available
ROASBenchROASBench: overall score 37.0 26th of 35 Claude Fable 5.2 61.3
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 37th of 159 available · 5th of 31 at its price
BenchmarkMistral Large 4Rank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 24.7% 25th of 107 Kimi K3.1 Pro 67.2%
  • tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
  • Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
  • Vending-Bench 2: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects PendingReleased after these sources last updated
  • APEX-Agents: pending. Its latest results are dated 1 Oct 2026, before this model was released.
  • GDP.pdf: pending. Its latest results are dated 1 Oct 2026, before this model was released.
  • Remote Labor Index: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Reasoning and knowledgeHard questions across science, maths and general knowledge 54th of 187 available · 11th of 35 at its price
BenchmarkMistral Large 4Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 38.4 34th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 35.0% 61st of 186 Claude Opus 5.5 61.4%
WritingWhat people prefer in blind comparisons, and judged writing quality PendingReleased after these sources last updated
  • Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
  • UGI Leaderboard: pending. Its latest results are dated 2 Oct 2026, before this model was released.
Sticking to the factsSummarising without inventing things, and factual answers PendingReleased after these sources last updated
  • Vectara Hallucination Leaderboard: pending. Its latest results are dated 22 Sep 2026, before this model was released.
  • Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
  • SimpleQA Verified (Epoch AI): pending. Its latest results are dated 29 Sep 2026, before this model was released.
Speed under pressureGood decisions against a real clock (fast chess) 27th of 38 available · 4th of 7 at its price
BenchmarkMistral Large 4Rank among availableBest available
BulletBenchBulletBench Lightning 10+1: ladder Elo 249 13th of 18 Gemini 3.5 Flash-Lite 892

Nearest alternatives

  • Cheaper at similar intelligenceMiMo-V2.6-Flash$0.18 blended · Intelligence Index 37.9

Against Mistral Large 4 at $2.06 blended, Intelligence Index 38.4. "Similar intelligence" means within 4 points or better.

Where to run it

2 hosts on OpenRouter at the standard service tier, with a median of $1.10 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Mistralnot stated$0.70$2.10$1.05
Mistralnot stated$0.77$2.31$1.16

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Mistral Large 4Muse Spark 1.3
most intelligent at $1–3 per million tokens
MiMo-V2.6-Flash
cheapest with similar intelligence
DeepSeek-V4.1-Flash
fastest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $2.06 $2.00$0.18$0.52$8.00
Tokens a second 60 2544636996
Intelligence Index 38.4 48.137.939.557.6
Spring Prompt overall 24 58–4980
Rank among available models, by use case
Product listings 30th of 35 20th–18th16th
Decks from an analysis 33rd of 35 28th––8th
User surveys 19th of 26 –––5th
Marketing planning 26th of 35 28th––11th
Agents and tool use 37th of 159 13th42nd31st3rd
Professional work pending 8th––3rd
Reasoning and knowledge 54th of 187 11th55th42nd1st
Writing pending 7th56th29th5th
Sticking to the facts pending 11th46th23rd3rd
Speed under pressure 27th of 38 36th–––

Shaded figures are better than Mistral Large 4's. A dash means no result.

Research and examples

In the playground: our pick in 0 of 20 tests, 79% of checks passed →

Where it has been measured

Most intelligent of the 15 models from developers based in the EU (all Mistral AI's). 5th most intelligent of the 34 models priced $1–3 per million tokens. Among models available today, in the top quarter (by percentile) for agents and tool use; in the bottom quarter for product listings and decks from an analysis. Results for professional work, writing and sticking to the facts are pending: those sources last updated before this model was released.

Every result

135 published results from 6 sources, each in the source's own units, with the configuration that produced it.

BulletBench · measured by Spring Prompt · 10 results
MeasureValueRankConfigurationDated
BulletBench Blitz 3+2: cost per gameUS dollars, lower is better $0.0082 2 of 25 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Blitz 3+2: games lost on time% of games, lower is better 0.0% 1 of 25 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Blitz 3+2: invalid moves% of moves, lower is better 0.6% 24 of 25 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Blitz 3+2: ladder Eloladder Elo, higher is better 555
392–727
20 of 25 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Blitz 3+2: median move timemilliseconds, lower is better 0.8 s 3 of 25 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Lightning 10+1: cost per gameUS dollars, lower is better $0.0117 40 of 48 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Lightning 10+1: games lost on time% of games, lower is better 58.3% 37 of 48 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Lightning 10+1: invalid moves% of moves, lower is better 1.8% 46 of 48 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Lightning 10+1: ladder Eloladder Elo, higher is better 249
0–486
40 of 48 Mistral Large 4
provider default reasoning
6 Oct 2026
BulletBench Lightning 10+1: median move timemilliseconds, lower is better 0.8 s 31 of 48 Mistral Large 4
provider default reasoning
6 Oct 2026

Not shown: Not your cost: prices are those charged on the run date.

CatalogBench · measured by Spring Prompt · 62 results
MeasureValueRankConfigurationDated
CatalogBench (sales brief): channel rules broken% of products, lower is better 1.2% 28 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): channel rules broken% of products, lower is better 4.2% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): claims to check% of products, lower is better 31.6% 34 of 35 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): claims to check% of products, lower is better 3.3% 27 of 35 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): content quality% of checks, higher is better 95.9% 29 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): content quality% of checks, higher is better 89.2% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): failed outputs% of products, lower is better 0.0% 1 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): failed outputs% of products, lower is better 7.7% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): missing UK information% of products, lower is better 0.0% 1 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): missing UK information% of products, lower is better 0.0% 1 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): not findable% of products, lower is better 6.0% 27 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): not findable% of products, lower is better 3.0% 15 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): publish-ready listings% of products, higher is better 2.4% 34 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): publish-ready listings% of products, higher is better 10.7% 29 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): reliably publish-ready% of products, higher is better 0.0% 30 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): reliably publish-ready% of products, higher is better 1.8% 28 of 37 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): unsupported claims% of products, lower is better 97.0% 38 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): unsupported claims% of products, lower is better 80.4% 31 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): unsupported claimsclaims per product, lower is better 6.74 38 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): unsupported claimsclaims per product, lower is better 3.69 32 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): wrong attributes% of products, lower is better 42.3% 37 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): wrong attributes% of products, lower is better 17.3% 20 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench (sales brief): wrong category or variant% of products, lower is better 1.8% 7 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench (sales brief): wrong category or variant% of products, lower is better 2.4% 21 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: channel compliance% of products, higher is better 100.0% 1 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: channel compliance% of products, higher is better 98.1% 38 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: channel rules broken% of products, lower is better 3.0% 36 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: channel rules broken% of products, lower is better 6.6% 38 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: claims to check% of products, lower is better 12.7% 34 of 35 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: claims to check% of products, lower is better 0.7% 27 of 35 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: conflicts caught% of conflicts, higher is better 100.0% 1 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: conflicts caught% of conflicts, higher is better 100.0% 1 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: content quality% of checks, higher is better 96.5% 23 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: content quality% of checks, higher is better 90.6% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: cost per productUS dollars, lower is better $0.0033 9 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: cost per productUS dollars, lower is better $0.0125 22 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: decision accuracy% of decisions, higher is better 93.1% 34 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: decision accuracy% of decisions, higher is better 88.7% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: failed outputs% of products, lower is better 0.0% 1 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: failed outputs% of products, lower is better 7.1% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: field accuracy% of missing fields, higher is better 94.4% 10 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: field accuracy% of missing fields, higher is better 86.9% 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: invented values% of filled values, lower is better 3.5% 32 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: invented values% of filled values, lower is better 4.4% 35 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: missing UK information% of products, lower is better 0.0% 1 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: missing UK information% of products, lower is better 0.0% 1 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: not findable% of products, lower is better 4.2% 18 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: not findable% of products, lower is better 5.4% 21 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: publish-ready listings% of products, higher is better 18.5% 33 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: publish-ready listings% of products, higher is better 39.9% 30 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: reliably publish-ready% of products, higher is better 7.1% 32 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: reliably publish-ready% of products, higher is better 10.7% 31 of 37 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: time per productseconds, lower is better 8.20 8 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: time per productseconds, lower is better 97.5 37 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: unsupported claims% of products, lower is better 64.3% 37 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: unsupported claims% of products, lower is better 33.3% 30 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: unsupported claimsclaims per product, lower is better 2.07 36 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: unsupported claimsclaims per product, lower is better 0.91 27 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: wrong attributes% of products, lower is better 38.1% 37 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: wrong attributes% of products, lower is better 19.6% 26 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
CatalogBench: wrong category or variant% of products, lower is better 1.2% 5 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
CatalogBench: wrong category or variant% of products, lower is better 4.2% 32 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026

Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.

DeckBench · measured by Spring Prompt · 17 results
MeasureValueRankConfigurationDated
DeckBench: accurate decks% of tasks, higher is better 0.0% 32 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: caveat dropped% of tasks, lower is better 16.7% 16 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: clean layout% of tasks, higher is better 33.3% 18 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: cost per deckUS dollars, lower is better $0.0148 9 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: deck ratingrating, higher is better 595
472–693
35 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: design quality% of the maximum, higher is better 30.5% 36 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: draft figure quoted% of tasks, lower is better 16.7% 15 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: findings missing% of tasks, lower is better 0.0% 1 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: head-to-head win rate% of comparisons, higher is better 24.7% 36 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: layout defects% of tasks, lower is better 66.7% 18 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: misleading metric used% of tasks, lower is better 0.0% 1 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: presentable decks% of tasks, higher is better 0.0% 17 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: recommendation late or wrong% of tasks, lower is better 83.3% 34 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: slides needing work% of tasks, lower is better 100.0% 37 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: time per deckseconds, lower is better 114 18 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: unsupported claims% of tasks, lower is better 50.0% 32 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026
DeckBench: unsupported numbers% of tasks, lower is better 33.3% 34 of 37 Mistral Large 4
provider default reasoning
6 Oct 2026

Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.

ROASBench · measured by Spring Prompt · 18 results
MeasureValueRankConfigurationDated
ROASBench: audience scorescore out of 100, higher is better 55.8 26 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: audience scorescore out of 100, higher is better 51.3 30 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: business scorescore out of 100, higher is better 34.3 26 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: business scorescore out of 100, higher is better 19.3 31 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: consistency scorescore out of 100, higher is better 16.7 30 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: consistency scorescore out of 100, higher is better 9.7 35 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: contribution profitsimulated US dollars, higher is better $533,401 24 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: contribution profitsimulated US dollars, higher is better $156,831 31 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: cost of a runUS dollars, lower is better $0.0657 10 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: cost of a runUS dollars, lower is better $0.19 18 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: months over budgetmonths of 12, lower is better 3 37 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: months over budgetmonths of 12, lower is better 0 1 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: overall scorescore out of 100, higher is better 37.0 26 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: overall scorescore out of 100, higher is better 27.4 31 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: planning scorescore out of 100, higher is better 54.4 29 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: planning scorescore out of 100, higher is better 55.0 7 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026
ROASBench: return on ad spendprofit per $1 spent, higher is better $0.45 23 of 38 Mistral Large 4
provider default reasoning
6 Oct 2026
ROASBench: return on ad spendprofit per $1 spent, higher is better $0.14 31 of 38 Mistral Large 4 (high reasoning)
high reasoning
6 Oct 2026

Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.

SurveyBench · measured by Spring Prompt · 20 results
MeasureValueRankConfigurationDated
SurveyBench: analysis ratingrating, higher is better 770
716–811
19 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: analysis scorepoints of 100, higher is better 63.8 19 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: cost per taskUS dollars, lower is better $0.0924 13 of 27 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: head-to-head win rate% of comparisons, higher is better 35.5% 21 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: leading or double-barrelled questions% of tasks, lower is better 33.3% 12 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: mistaken next steps% of tasks, lower is better 66.7% 23 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: numbers right% of answers, higher is better 98.9% 19 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: objectives unanswerable% of tasks, lower is better 11.1% 18 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: precision ignored% of tasks, lower is better 55.6% 23 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: research scorepoints of 100, higher is better 71.2 19 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: sound research% of tasks, higher is better 0.0% 17 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: stakeholder's view presumed% of tasks, lower is better 22.2% 18 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: survey plan scorepoints of 100, higher is better 88.4 19 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: survey structure% of tasks, lower is better 11.1% 3 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: time per taskseconds, lower is better 439 25 of 27 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: traps handled% of traps, higher is better 78.7% 20 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: traps missed% of tasks, lower is better 100.0% 25 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: unsupported findings% of tasks, lower is better 100.0% 24 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: wrong numbers% of tasks, lower is better 22.2% 17 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026
SurveyBench: wrong recommendation% of tasks, lower is better 44.4% 19 of 28 Mistral Large 4
provider default reasoning
8 Oct 2026

Not shown: Not your customers or your data: invented organisations with simulated respondents, nine public tasks, one run per model per task at the provider's default reasoning setting.

Artificial Analysis · reported by Artificial Analysis · 8 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 38.4 73 of 413 Mistral Large 4 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 35.0% 121 of 419 Mistral Large 4 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 81.3% 53 of 407 Mistral Large 4 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 60.2 100 of 137 Mistral Large 4 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 54.2% 65 of 205 Mistral Large 4 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 24.7% 61 of 204 Mistral Large 4 9 Nov 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 23.2% 62 of 204 Mistral Large 4 (medium reasoning) 9 Nov 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 33.9 100 of 137 Mistral Large 4 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

Embed a score badge

A badge for its Spring Prompt Score and each of our benchmarks that ranked Mistral Large 4, from the current results. Paste the code into a page or README; it links back to where the result comes from.

Mistral Large 4 is #33 of 35 on the Spring Prompt Score (22 of 100, from 5 of 5 of our benchmarks), measured by Spring Prompt Spring Prompt Score · 5 of 5 benchmarks
<a href="https://springprompt.com/score#mistral-mistral-large-4-0"><img src="https://springprompt.com/badge/score/mistral-mistral-large-4-0.svg" alt="Mistral Large 4 is #33 of 35 on the Spring Prompt Score (22 of 100, from 5 of 5 of our benchmarks), measured by Spring Prompt" height="20"></a>
Mistral Large 4 is #31 of 35 on CatalogBench (Reliably publish-ready, factual brief: 10.7%), measured by Spring Prompt CatalogBench · Reliably publish-ready, factual brief
<a href="https://springprompt.com/benchmarks/catalogbench?focus=mistral-mistral-large-4-0"><img src="https://springprompt.com/badge/catalogbench/mistral-mistral-large-4-0.svg" alt="Mistral Large 4 is #31 of 35 on CatalogBench (Reliably publish-ready, factual brief: 10.7%), measured by Spring Prompt" height="20"></a>
Mistral Large 4 is #34 of 36 on DeckBench (Deck rating: 595), measured by Spring Prompt DeckBench · Deck rating
<a href="https://springprompt.com/benchmarks/deckbench?focus=mistral-mistral-large-4-0"><img src="https://springprompt.com/badge/deckbench/mistral-mistral-large-4-0.svg" alt="Mistral Large 4 is #34 of 36 on DeckBench (Deck rating: 595), measured by Spring Prompt" height="20"></a>
Mistral Large 4 is #19 of 27 on SurveyBench (Analysis rating: 770), measured by Spring Prompt SurveyBench · Analysis rating
<a href="https://springprompt.com/benchmarks/surveybench?focus=mistral-mistral-large-4-0"><img src="https://springprompt.com/badge/surveybench/mistral-mistral-large-4-0.svg" alt="Mistral Large 4 is #19 of 27 on SurveyBench (Analysis rating: 770), measured by Spring Prompt" height="20"></a>
Mistral Large 4 is #26 of 36 on ROASBench (Score: 37.0), measured by Spring Prompt ROASBench · Score
<a href="https://springprompt.com/benchmarks/roasbench?focus=mistral-mistral-large-4-0"><img src="https://springprompt.com/badge/roasbench/mistral-mistral-large-4-0.svg" alt="Mistral Large 4 is #26 of 36 on ROASBench (Score: 37.0), measured by Spring Prompt" height="20"></a>
Mistral Large 4 is #39 of 45 on BulletBench (Lightning 10+1: 249), measured by Spring Prompt BulletBench · Lightning 10+1
<a href="https://springprompt.com/benchmarks/bulletbench?focus=mistral-mistral-large-4-0"><img src="https://springprompt.com/badge/bulletbench/mistral-mistral-large-4-0.svg" alt="Mistral Large 4 is #39 of 45 on BulletBench (Lightning 10+1: 249), measured by Spring Prompt" height="20"></a>

Will Mistral Large 4 work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.