Models / Llama 4 Maverick 17B 128E Instruct (FP8)

Meta

Llama 4 Maverick 17B 128E Instruct (FP8)

Early evidence, 10 results from 2 sources, on 2 use cases. Llama 4 Maverick 17B 128E Instruct (FP8) is strongest at sticking to the facts; it is not on sale through OpenRouter, so it is here for reference.

Results 16 Nov 2026 · 10 results · 2 sources

Price per million tokens
Not available through OpenRouter
Speed
Not measured
Intelligence Index
Not measured
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Meta, based in the US
Weights not published
Released
Not recorded
no launch date in our tracker or our sources

How good is it, and for what?

Each use case ranks the available models its benchmarks measured. 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 Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 86th of 160 available
BenchmarkLlama 4 Maverick 17B 128E Instruct (FP8)Rank among availableBest available
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 37.3% 20th of 37 Claude Opus 4.5 77.5%
  • tau2-bench: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge Not measured
  • Artificial Analysis: has not measured this model.
WritingWhat people prefer in blind comparisons, and judged writing quality Not measured
  • Arena (formerly LMArena): has not measured this model.
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 50th of 140 available
BenchmarkLlama 4 Maverick 17B 128E Instruct (FP8)Rank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 8.2% 28th of 72 GPT-5.4 nano 3.1%
  • Arena (formerly LMArena): has not measured this model.
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

  • The most capable open-weights modelMiMo-V2.6-Pro$0.54 blended · Intelligence Index 46.3

Against Llama 4 Maverick 17B 128E Instruct (FP8). "Similar intelligence" means within 4 points or better.

Against the alternatives

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

Scroll sideways to see every alternative.

Llama 4 Maverick 17B 128E Instruct (FP8)Claude Fable 5.2
leads overall
Muse Wonder
Meta's best other model
Claude Opus 5.5
near the top overall
GPT-6.1 Sol
near the top overall
Price per million tokens – $16.00$8.00$8.00$4.00
Tokens a second – –1539653
Intelligence Index – –52.457.651.8
Spring Prompt overall – 90708280
Rank among available models, by use case
Product listings – 5th5th13th2nd
Decks from an analysis – 4th13th7th5th
User surveys – 1st2nd4th3rd
Marketing planning – 1st22nd9th–
Agents and tool use 86th of 160 –11th3rd6th
Professional work – ––3rd9th
Reasoning and knowledge – –8th1st6th
Writing – ––5th16th
Sticking to the facts 50th of 140 ––3rd20th
Speed under pressure – 8th14th–3rd

Shaded figures are better than Llama 4 Maverick 17B 128E Instruct (FP8)'s. A dash means no result.

Where it has been measured

Compared with the models available today (it is not on sale itself), mid-field on agents and tool use, and sticking to the facts, the use cases measured so far.

Not timed: Artificial Analysis, our source for speed, has not tested it

Every result

10 published results from 2 sources, each in the source's own units, with the configuration that produced it.

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 8 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 56.0% 94 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 18.9% 47 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 20.3% 52 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 37.3% 50 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 100.0% 1 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 88.7% 13 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 73.7% 49 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 28.0% 35 of 109 Llama 4 Maverick 17B 128E Instruct (FP8, bfcl fc) 16 Dec 2025

Not shown: Not general refusal or safety behaviour.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 100.0% 1 of 108 Llama 4 Maverick 17B 128E Instruct (FP8) 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 8.2% 40 of 108 Llama 4 Maverick 17B 128E Instruct (FP8) 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Test your own prompt.

Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.