Compare / Meta vs Mistral AI

AI labs compared · November 2026

Meta vs Mistral AI: which AI lab is ahead?

Each lab's best model on sale goes head to head on every kind of work we measure, from our own benchmarks and licensed sources. 23 models in all; a lab is only as good as the model you would actually pick from it.

11 models on sale with our resultsMeta Best: Muse Wonder · newest: Muse Ember
12 models on sale with our resultsMistral AI Best: Mistral Large 4 · newest: Mistral Large 4
Meta wins 7Mistral AI wins 2

Kinds of work led by each lab's best model there, of the 9 both are measured on.

Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331). Models on sale means listed on OpenRouter today.

Meta or Mistral AI? The short answer

Meta is ahead: its best models beat Mistral AI's in 7 of the 9 kinds of work both are measured on; Mistral AI leads in 2.

  • Pick Meta for product listings, decks from an analysis, user surveys, marketing planning, agents and tool use, reasoning and knowledge and writing.
  • Pick Mistral AI for sticking to the facts and speed under pressure.
  • On our overall leaderboard, Muse Wonder is #10 (66 of 100) and Mistral Large 4 is #35 (24).
  • Prices start at $0.058 per million tokens for Meta (Llama 3.1 8B Instruct) and $0.058 for Mistral AI (Mistral Small 3).

Head to head

66Muse Wonder
Best Spring Prompt overall score
24Mistral Large 4
52.4Muse Wonder
Best Intelligence Index
38.4Mistral Large 4
$8.00Muse Wonder
Flagship price, per million tokens
$2.06Mistral Large 4
$0.058Llama 3.1 8B Instruct
Cheapest model, per million tokens
$0.058Mistral Small 3
11
Models on sale with our results
12

Each line names the model behind the number. Prices are blended, three input tokens to one output, at the developer's list price where we have read it; see every model's price. Intelligence Index from Artificial Analysis; overall score from our leaderboard.

Which lab is better for what

For each kind of work, each lab's best model there and its rank among every model on sale. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live 8 of 35Muse Wonder 26 of 35Mistral Medium 3.5
Decks from an analysisTurning a finished analysis into a deck you could present as it is 15 of 35Muse Wonder 33 of 35Mistral Large 4
User surveysPlanning a user survey and reading its results without being misled 3 of 26Muse Wonder 19 of 26Mistral Large 4
Marketing planningPlanning a year of ad spend without overspending 10 of 35Muse Ember 26 of 35Mistral Large 4
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 11 of 159Muse Wonder 37 of 159Mistral Large 4
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 8 of 51Muse Spark 1.3 not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge 8 of 187Muse Wonder 54 of 187Mistral Large 4
WritingWhat people prefer in blind comparisons, and judged writing quality 7 of 169Muse Spark 1.3 85 of 169Mistral Medium 3.5
Sticking to the factsSummarising without inventing things, and factual answers 11 of 139Muse Spark 1.3 8 of 139Mistral Small 3
Speed under pressureGood decisions against a real clock (fast chess) 17 of 38Muse Wonder 13 of 38Mistral Medium 3.5

Every model, by intelligence and price

Each dot is a model on sale with an Intelligence Index from Artificial Analysis. Up is smarter; left is cheaper. Hover or tap a dot for its name.

MetaMistral AI

0204060 $0.01$0.1$1$10 Price per million tokens, blended (log scale) Intelligence Index Muse WonderMuse Spark 1.3Muse Spark 1.2Muse Spark 1.1Mistral Large 4Mistral Medium 3.5

On our overall leaderboard

The 5 of their models we rank overall: one score out of 100 across everything we publish (how the score works). Only models with enough of our own results are ranked.

  1. #10Muse Wonder66
  2. #13Muse Ember61
  3. #17Muse Spark 1.358
  4. #35Mistral Large 424
  5. #37Mistral Medium 3.515

Benchmark by benchmark

The headline result of every benchmark both labs have models on: each lab's best model and its rank among the models on sale. Green marks the better rank.

MeasureMetaMistral AI
Agents and tool use
Berkeley Function Calling Leaderboard (BFCL) V4 · Overall accuracy · BFCL% correct, higher is better 31.9%Llama 3.3 70B Instruct · 24 of 3637.7%Mistral Medium 3 · 19 of 36
Artificial Analysis · Terminal-Bench 4.0% of tasks, higher is better 52.0%Muse Wonder · 9 of 10724.7%Mistral Large 4 · 25 of 107
Artificial Analysis · Terminal-Bench 2.1% of tasks, higher is better 85.4%Muse Spark 1.3 · 10 of 10750.6%Mistral Medium 3.5 · 59 of 107
Artificial Analysis · τ-bench banking% of tasks, higher is better 50.5%Muse Spark 1.3 · 2 of 10415.1%Mistral Medium 3.5 · 56 of 104
Decks from an analysis
DeckBench · Deck ratingrating, higher is better 1,204Muse Wonder · 15 of 35595Mistral Large 4 · 33 of 35
Marketing planning
ROASBench · Overall scorescore out of 100, higher is better 51.8Muse Ember · 10 of 3537.0Mistral Large 4 · 26 of 35
Product listings
CatalogBench · Reliably publish-ready% of products, higher is better 66.1%Muse Wonder · 7 of 3517.9%Mistral Medium 3.5 · 29 of 35
CatalogBench · Reliably publish-ready · sales brief% of products, higher is better 51.8%Muse Wonder · 7 of 3510.7%Mistral Medium 3.5 · 20 of 35
Reasoning and knowledge
Artificial Analysis · Artificial Analysis Intelligence Indexindex score, higher is better 52.4Muse Wonder · 6 of 18538.4Mistral Large 4 · 34 of 185
Artificial Analysis · Humanity's Last Exam% of questions, higher is better 50.1%Muse Wonder · 10 of 18635.0%Mistral Large 4 · 61 of 186
Artificial Analysis · GPQA Diamond% of questions, higher is better 94.1%Muse Spark 1.3 · 5 of 16876.9%Mistral Small 4 · 104 of 168
Speed under pressure
BulletBench · Ladder Elo · Bullet 60sladder Elo, higher is better 409Muse Wonder · 18 of 36572Mistral Medium 3.5 · 9 of 36
Sticking to the facts
Vectara Hallucination Leaderboard · Hallucination rate · Vectara% of summaries, lower is better 7.7%Llama 4 Scout · 26 of 715.1%Mistral Small 3 · 5 of 71
Arena (formerly LMArena) · Overall · Text factualityArena rating, higher is better 1,483Muse Spark 1.2 · 10 of 1091,434Mistral Large 3 · 72 of 109
User surveys
SurveyBench · Analysis ratingrating, higher is better 1,439Muse Wonder · 3 of 26770Mistral Large 4 · 19 of 26
Writing
Arena (formerly LMArena) · Overall · TextArena rating, higher is better 1,494Muse Spark 1.3 · 8 of 1531,427Mistral Medium 3.5 · 74 of 153
UGI Leaderboard · Writing score · UGIscore out of 100, higher is better 26.7Llama 4 Scout · 110 of 12445.5Mistral Medium 3.5 · 64 of 124

Quick answers

Is Meta or Mistral AI better?

Meta is ahead: its best models beat Mistral AI's in 7 of the 9 kinds of work both are measured on; Mistral AI leads in 2.

What is Meta's best model?

Muse Wonder: #10 on the Spring Prompt overall leaderboard and 52.4 on the Artificial Analysis Intelligence Index.

What is Mistral AI's best model?

Mistral Large 4: #35 on the Spring Prompt overall leaderboard and 38.4 on the Artificial Analysis Intelligence Index.

Which is better for agents and tool use, Meta or Mistral AI?

Meta: its best model there, Muse Wonder, ranks 11 of 159 models on sale, against 37 for Mistral AI's Mistral Large 4.

Which is better for reasoning and knowledge, Meta or Mistral AI?

Meta: its best model there, Muse Wonder, ranks 8 of 187 models on sale, against 54 for Mistral AI's Mistral Large 4.

Which is better for writing, Meta or Mistral AI?

Meta: its best model there, Muse Spark 1.3, ranks 7 of 169 models on sale, against 85 for Mistral AI's Mistral Medium 3.5.

Which is cheaper, Meta or Mistral AI?

Mistral AI's cheapest model with results, Mistral Small 3, costs $0.058 per million tokens (three input to one output), against $0.058 for Llama 3.1 8B Instruct from Meta.

Other labs compared

Which model is best for your product?

Benchmarks aren't your task. Run models from Meta, Mistral AI and the rest on your own prompt, and see the cheapest one that passes.