Llama 4 Maverick
Meta · released Apr 1, 2025 · meta-llama/Llama-4-Maverick-17B-128E-Instruct
- Type
- Open weightsCustom licence
- Params
- 402B
- Context
- 1M
about 786K words of context · download allowed, licence restricts use
Our take
Written Aug 2, 2026Llama 4 Maverick is a large downloadable model from Meta that accepts text and images and can handle up to one million tokens in a single request. It is a strong pick for coding workloads and long-document analysis, though its custom licence is less permissive than some alternatives.
Choose this for coding tasks or long-context document work at up to one million tokens. It suits budget-conscious hosted inference with a wide speed-to-price trade-off. Skip it if you need a permissive Apache-style licence, if creative writing is your main workload, or if you want a verified active parameter count.
The case for it
- Extremely large request limit for a downloadable model: 1,048,576 tokens.
- Strongest measured performance in coding among its own skill profiles, 65.93 points above its own creative-writing score.
- Wide price range lets users trade cost for speed: 2.5× throughput spread for under twice the output price increase.
- Broad multimodal input, accepting both text and images.
The case against it
- Custom licence, less permissive than Apache 2.0; commercial restrictions are possible but not independently verified.
- Creative writing is a relative weak spot versus its own other skills, 65.93 points below its coding score.
- Active parameter count undisclosed, so efficiency claims are unverifiable.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)112th of 143 · 1327.2
CodingWriting and fixing code on its own
Arena Coding109th of 143 · 1373.3
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Llama 4 Maverick is not on Arena Agent (IPS), which is where the rating would come from, so there is no rating here. It is on SWE-bench Verified, in 37th of 39 with 21.
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Llama 4 Maverick placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model7 scoresEvery figure we hold, from 7 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Comfortable fit
Apple M3 Ultra (80-core GPU) · 512 GB
Room to spare. 121.2 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 8 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.20 in / $0.80 out
- Context served
- 1M
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| DeepInfrafp8 | $0.20 / $0.80 | 1M | 33 tok/s | No | No | Unknown |
| OpenRouter | $0.20 / $0.80 | 1M | not measured | Unknown | Unknown | Unknown |
| DeepInfrabasefp8 | $0.20 / $0.80 | 1M | 24 tok/s | No | No | Unknown |
| Novita AIfp8 | $0.27 / $0.85 | 1M | 29 tok/s | No | No | Confirmed |
| Novita AI | $0.27 / $0.85 | 1M | not measured | Unknown | Unknown | Unknown |
| DigitalOcean Gradient | $0.25 / $0.87 | 128K | 17 tok/s | No | No | Confirmed |
| Parasailfp8 | $0.35 / $1.00 | 524K | 57 tok/s | No | No | Confirmed |
| Google Vertex AIus-east5 | $0.35 / $1.15 | 524K | 40 tok/s | No | No | Confirmed |
DeepInfra
Novita AI
DigitalOcean Gradient
Google Vertex AIAcross the 8 listings we hold: 6 say they do not train on prompts, 0 say they do and 2 do not say. 4 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| DeepInfrafp8 | ✗ | ✓ | ✓ |
| OpenRouter | ✓ | ✓ | ✓ |
| DeepInfrabase · fp8 | ✗ | ✓ | ✓ |
| Novita AIfp8 | ✗ | ✓ | ✓ |
| Novita AI | |||
| DigitalOcean Gradient | ✓ | ✓ | ✓ |
| Parasailfp8 | ✓ | ✓ | ✓ |
| Google Vertex AIus-east5 | ✓ | ✓ | ✓ |
Tool calling: 4 of 8 listings say yes, 3 say no, 1 publishes no parameter list. JSON output: 7 of 8 listings say yes, 1 publishes no parameter list. Strict schema: 7 of 8 listings say yes, 1 publishes no parameter list.
Models people weigh against Llama 4 Maverick
When we formed this view
Dates behind this page
Prices last checked 14h ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- 1 of 8 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 8 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsCustom licence, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Custom licence
This model ships custom license terms that don't map to a known template. We haven't parsed them, so commercial use, redistribution and derivatives are unverified — review the original terms before shipping.
Identifiers
- Hugging Face
- meta-llama/Llama-4-Maverick-17B-128E-Instruct
- Modality record
- text+image->text
- Catalogue slug
- meta-llama-llama-4-maverick