Llama-4-Maverick-17B-128E-Instruct
Access to model meta-llama/Llama-4-Maverick-17B-128E-Instruct is restricted.
text + image → text · made by Meta Llama
Sold by 5 ways
| Seller | Lane | Rate |
|---|---|---|
| Google Vertex AI cloud | standard | $0.35per Mtok in$1.15per Mtok outmodels.dev · read 2026-11-01 |
| ElectronHub aggregator | standard | $0.2per Mtok in$0.8per Mtok outapi.electronhub.ai · read 2026-10-26 |
| Groq 0 ways | $0.2per Mtok in$0.6per Mtok out | |
| Groq aggregatornot on the seller's list today | standard | $0.2per Mtok in$0.6per Mtok outraw.githubusercontent.com · read 2026-09-22 |
| IO.NET aggregator | standard | $0.27per Mtok in$0.83per Mtok out$0.13per Mtok cached$0.3per Mtok cache writemodels.dev · read 2026-11-01 |
| SambaNova Cloud aggregator | standard | $0.63per Mtok in$1.8per Mtok outraw.githubusercontent.com · read 2026-11-01 |
Measured 14 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 8thof 8 | Spatialviz bench — Epoch AI | Overall score | 0.318 |
| 35thof 42 | Fictionlivebench — Epoch AI | 120k token score | 0.364 |
| 45thof 46 | Enigma eval — Epoch AI | Accuracy | 0.58 |
| 46thof 49 | Lech mazur writing — Epoch AI | Mean score | 6.2 |
| 47thof 56 | Humanity's Last Exam (Epoch AI replication) | Accuracy | 5.68 |
| 64thof 72 | Aider polyglot — Epoch AI | Percent correct | 15.6 |
| 83rdof 102 | Simplebench — Epoch AI | Score (AVG@5) | 0.277 |
| 155thof 173 | Weirdml — Epoch AI | Accuracy | 24.47 |
| 173rdof 197 | Lmca — Epoch AI | Score | 15.87 |
| 183rdof 235 | Dtbench — Epoch AI | Accuracy | 61.87 |
| 200thof 260 | Critpt — Epoch AI | Accuracy | 0 |
| 215thof 235 | Scicode — Epoch AI | Score | 31.713 |
| 276thof 290 | Arc agi 2 — Epoch AI | Score | 0 |
| 304thof 309 | Arc agi — Epoch AI | Score | 4.38 |
About
Llama-4-Maverick-17B-128E-Instruct — a text model from Meta Llama, sold by 4 companies from $0.2 in and $0.8 out per million tokens, placed 200th of 260 on Critpt — Epoch AI.
It takes text and images and returns text, with a context window of 524,288 tokens. It was published in April 2025. Its sellers say it can call a tool. The catalogue files it under reasoning. Its weights are published under the other licence, which attaches conditions the plain open licences do not. Four companies sell it. The cheapest is $0.2 in and $0.8 out per million tokens at ElectronHub. It has been measured on 14 boards, and stands best at 200th of 260 on Critpt — Epoch AI.
Every current figure
- Maker
- Meta Llama
- Register
- model
- Takes
- text + image
- Returns
- text
- Context
- 524,288 tokens
- Longest answer
- 8,192 tokens
- Published
- April 2025
- Parameters
- 402 billion · read from its own weights
- Licence
- other
- Sellers
- 4
- Maker's own price
- not read
- Price
- $0.2 in and $0.8 out per million tokens — ElectronHub
- Boards
- 14
- Best place
- 200th of 260 — Critpt — Epoch AI
Known as 1 name
meta-llama/Llama-4-Maverick-17B-128E-Instruct