AI Model Watch
Models / Meta

Llama 4 Maverick (17B-128E Instruct)

GA

Open-weight, natively multimodal MoE: 17B active / 400B total params, 128 experts. License: Llama 4 Community License Agreement (commercial use permitted for orgs with <700M MAU). Meta is the model owner; no first-party Meta API pricing for self-host. Official Llama API model ID is 'Llama-4-Maverick-17B-128E-Instruct-FP8' and the Llama API serves it at a 128k context window (developer.meta.com/Llama API docs), whereas the open weights support up to 1M tokens. Hosted price shown is OpenRouter slug 'meta-llama/llama-4-maverick' = $0.15 in / $0.60 out per 1M (OpenRouter page accessed 2026-06-28).

Llama 4 Maverick (17B-128E Instruct) by Meta costs $0.15 per 1M input tokens and $0.6 per 1M output tokens ($0.262/1M blended), with a 1M (1.000.000-token) context window. It is generally available (GA), with a knowledge cutoff of Aug 2024. Its weights are publicly downloadable, so it can be self-hosted on your own infrastructure.

Last verified: 13 Aug 2026 · sourced from official provider documentation

Provider
Meta
Status
GA
Input price
$0.15 / 1M tokens
Output price
$0.6 / 1M tokens
Cached input
—
Blended price
$0.262 / 1M tokens
Context window
1.000.000 tokens (1M)
Max output
—
Modality
text, image
Open weights
Yes — self-hostable
Knowledge cutoff
2024-08
Released
5 Apr 2025
API string
meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8

Source: Meta official documentation ↗

See also

Compare Llama 4 Maverick (17B-128E Instruct) head-to-head

Llama 4 Maverick (17B-128E Instruct) vs Claude Opus 4.8→

$0.262 vs $10 blended /M · cross-provider

Llama 4 Maverick (17B-128E Instruct) vs GPT-5.6 Sol→

$0.262 vs $11.25 blended /M · cross-provider

Llama 4 Maverick (17B-128E Instruct) vs Gemini 3.6 Flash→

$0.262 vs $3 blended /M · cross-provider

Llama 4 Maverick (17B-128E Instruct) vs Grok 4.5→

$0.262 vs $3 blended /M · cross-provider

Llama 4 Maverick (17B-128E Instruct) vs Mistral Large 3→

$0.262 vs $0.75 blended /M · cross-provider

Llama 4 Maverick (17B-128E Instruct) vs Amazon Nova Pro→

$0.262 vs $1.40 blended /M · cross-provider

FAQ

Llama 4 Maverick (17B-128E Instruct) — questions & answers

How much does Llama 4 Maverick (17B-128E Instruct) cost?

Llama 4 Maverick (17B-128E Instruct) is priced at $0.15 per 1M input tokens and $0.6 per 1M output tokens ($0.262/1M blended at a 3:1 input-to-output ratio).

What is the context window of Llama 4 Maverick (17B-128E Instruct)?

Llama 4 Maverick (17B-128E Instruct) has a 1.000.000-token context window (1M).

Is Llama 4 Maverick (17B-128E Instruct) deprecated?

No — Llama 4 Maverick (17B-128E Instruct) is generally available (GA) and not currently scheduled for deprecation or retirement in our tracker.

Does Llama 4 Maverick (17B-128E Instruct) support image or vision input?

Yes — Llama 4 Maverick (17B-128E Instruct) accepts image input. Listed modalities: text, image.

Can I self-host Llama 4 Maverick (17B-128E Instruct)?

Yes — Llama 4 Maverick (17B-128E Instruct) is an open-weight model: its weights are publicly downloadable, so you can run it on your own infrastructure (subject to its license) instead of only calling Meta's API.

What is the API model string for Llama 4 Maverick (17B-128E Instruct)?

The API model identifier for Llama 4 Maverick (17B-128E Instruct) is "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8" when calling Meta's API.

What is Llama 4 Maverick (17B-128E Instruct)'s knowledge cutoff?

Llama 4 Maverick (17B-128E Instruct)'s training knowledge cutoff is Aug 2024.