Llama 4 Maverick

Featured
Meta Llama 4 Generally Available Apr 2025

Meta's most capable open-weights model. 17B active / 400B total MoE with 1M context, strong MMLU scores that match GPT-4o, and native multimodal support.

Context
1M
tokens
Input
$0.15
per MTok
Output
$0.60
per MTok

About

Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...

Modalities

Input
Text Vision Code
Output
Text Code

Code Examples

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-17B-128E",
    "messages": [
      { "role": "user", "content": "Explain quantum entanglement in one sentence." }
    ]
  }'

API Parameters

Name Type Description
frequency_penalty number Penalize tokens by their frequency so far. Positive values reduce repetition.
logit_bias object Map of token-id to bias (-100…100) added to the logit before sampling.
max_tokens deprecated integer Deprecated. Use max_completion_tokens.
min_p unknown
presence_penalty number Penalize tokens that have appeared at all so far. Positive values encourage new topics.
repetition_penalty number Penalize repeated tokens (>1.0 reduces repetition, <1.0 encourages it).
response_format one of Constrain output to a JSON schema or an enum (structured outputs).
seed integer Deterministic seed for sampling. Same seed + same prompt produces identical output.
stop array Up to 4 sequences where the API will stop generating tokens.
structured_outputs boolean Enable JSON-schema-constrained output.
temperature number Sampling temperature; higher values produce more random output. 0 is deterministic.
top_k integer Limit sampling to the top-k most likely tokens at each step.
top_p number Nucleus sampling: consider only tokens whose cumulative probability ≥ top_p.

Standard OpenAI-compatible parameters. Consult the provider docs for model-specific behaviour.

Benchmark Scores

Benchmark Score
MMLU 85.5%
MMLU-Pro 80.5%
HumanEval 86%
MATH 73.5%

Strengths & Limitations

Best For
GPT-4o-level MMLU
1M token context
Open weights — self-hostable
Natively multimodal
Strong code generation
Limitations
Requires large GPU cluster for full deployment
Smaller HF community vs Scout at launch

Tags

Open WeightsMultimodalMoEReasoningVision