Skip to content
aiqbench

Meta · model passport

Meta: Llama 4 Maverick

llama-4-maverick-17b-128e-instruct
  • Open weights
  • Llama 4 licence
  • 401.6B params
  • 1,048,576 context
  • Released 2025-04-05(curated)
9/27axes covered
Overall · LMArena
1,327#229 of 396
Best percentile
52.2LiveBench
Cohort standing
38.0median 48.1

What the data says

Benchmarked on 9 of 27 tracked axes. Every figure below is the model's own published result on that board.

Weakest axes

  • 11.2 pct — ARC-AGI-2 (0.0%)

Uneven: 41 percentile points between its best and worst axis — 52.2 pct on LiveBench versus 11.2 pct on ARC-AGI-2.

  • LiveBench52.2 pct54.4%Verify →
  • Coding43.5 pct1,373Verify →
  • Hard prompts43.4 pct1,339Verify →
  • Overall42.4 pct1,327Verify →
  • Instruction following42.2 pct1,314Verify →
  • Vision Arena30 pct1,145Verify →
  • SimpleBench21.8 pct27.7%Verify →
  • Humanity's Last Exam19.6 pct5.7%Verify →
  • ARC-AGI-211.2 pct0.0%Verify →
  • WebDev ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

Repositorymeta-llama/Llama-4-Maverick-17B-128E-Instruct
Parameters401.6B
ArchitectureLlama4ForConditionalGeneration
LicenceLlama 4
AccessGated — you must accept the repo's terms before downloading

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "meta-llama/Llama-4-Maverick-17B-128E-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Use it hosted

API model IDmeta-llama/llama-4-maverick
Context length1,048,576 tokens
Max output16,384 tokens
Acceptstext, image
Input price$0.19 per 1M tokens
Cached input price$0.05 per 1M tokens
Output price$0.65 per 1M tokens

API parameters this endpoint supports

Frequency penaltyLogit biasLogprobsMax tokensPresence penaltyRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop logprobsTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"meta-llama/llama-4-maverick","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • llama-4-maverick-17b-128e-instruct stands at 38 (raw mean 34) — measured on 9 of 27 axes.

    standingtemplateVerify →

  • llama-4-maverick-17b-128e-instruct stands at 38 versus a cohort median of 48.1 across 151 models released within 180 days — below the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →