Model Library
MetaReleased April 5, 2025Llama 4 Maverick
Meta · Released April 5, 2025
Intelligence score
14
At a glance
- Intelligence
- 14
- Speed
- 118.1tokens/sec
- Latency
- 970mstime to first token
- Input price
- $0.35per 1M tokens
- Output price
- $0.85per 1M tokens
- Context window
- 4.0Mtokens
- Parameters
- 402billions
Summary
Comparison summary
Llama 4 Maverick scores 14 on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 17). Llama 4 Maverick generates output at 118.1 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight non-reasoning models of similar size (median: 58.3 t/s). Llama 4 Maverick costs $0.35 per 1M input tokens (very competitive, median: $0.60) and $0.85 per 1M output tokens (very competitive, median: $2.40), based on the median across providers serving the model.
Specifications
Technical specifications
| Model name | Llama 4 Maverick |
|---|---|
| Provider | Meta |
| Release date | April 5, 2025 |
| Intelligence score | 14 |
| Speed | 118.1 tokens/sec |
| Latency (TTFT) | 0.97s |
| Input price | $0.35 / 1M tokens |
| Output price | $0.85 / 1M tokens |
| Context window | 4,000,000 tokens |
| Parameters | 402B |
| Reasoning | No |
| Input modalities | text, image |
| Output modalities | — |
| Open weights | Yes |
| License | LLAMA 4 COMMUNITY LICENSE AGREEMENT |
| API availability | — |
| Source | Model Intelligence legacy profile: llama-4-maverick |
Pricing
Token prices
Price per 1M tokens
Llama 4 Maverick, as listed in the catalog record
- Input$0.35
- Output$0.85
| Token type | Price per 1M |
|---|---|
| Input | $0.35 |
| Output | $0.85 |
Model Intelligence legacy profile: llama-4-maverick
FAQ · 22
Frequently asked questions
Llama 4 Maverick was released on April 5, 2025.
Llama 4 Maverick was created by Meta.
Llama 4 Maverick scores 14 on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 17).
Llama 4 Maverick generates output at 118.1 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight non-reasoning models of similar size (median: 58.3 t/s).
Llama 4 Maverick has a time to first token (TTFT) of 0.97s (based on the median across providers serving the model), which is very competitive compared to other open weight non-reasoning models of similar size (median: 2.33s).
Llama 4 Maverick costs $0.35 per 1M input tokens (very competitive, median: $0.60) and $0.85 per 1M output tokens (very competitive, median: $2.40), based on the median across providers serving the model.
Llama 4 Maverick costs $0.35 per 1M input tokens and $0.85 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.34 per 1M tokens. Pricing may vary by provider.
When evaluated on the Intelligence Index, Llama 4 Maverick generated 4.8M output tokens, which is better than average compared to other open weight non-reasoning models of similar size (median: 4.8M).
No, Llama 4 Maverick is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Llama 4 Maverick supports text and image input.
Llama 4 Maverick supports text output.
Yes, Llama 4 Maverick supports image input and can analyze, describe, and answer questions about images.
Yes, Llama 4 Maverick is multimodal. It can process text and image input and generate text output.
Llama 4 Maverick has a context window of 1.0M tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Llama 4 Maverick is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Llama 4 Maverick has 402 billion parameters (17 billion active).
Llama 4 Maverick is a Mixture of Experts (MoE) model with 402 billion total parameters, but only 17 billion active parameters are used during inference.
Llama 4 Maverick is released under the LLAMA 4 COMMUNITY LICENSE AGREEMENT license. This license allows commercial use.
Llama 4 Maverick achieves a score of 14 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Llama 4 Maverick has a knowledge cutoff of August 2024. The model's training data includes information up to this date.
Yes, Llama 4 Maverick is available via API through 9 providers.
Llama 4 Maverick is available through 9 API providers.