Model Library
MetaReleased April 5, 2025Llama 4 Scout
Meta · Released April 5, 2025
Intelligence score
10
At a glance
- Intelligence
- 10
- Speed
- 108.5tokens/sec
- Latency
- 860mstime to first token
- Input price
- $0.17per 1M tokens
- Output price
- $0.66per 1M tokens
- Context window
- 10.0Mtokens
- Parameters
- 109billions
Summary
Comparison summary
Llama 4 Scout scores 10 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 7). Llama 4 Scout generates output at 108.5 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s). Llama 4 Scout costs $0.17 per 1M input tokens (very competitive, median: $0.53) and $0.66 per 1M output tokens (better than average, median: $1.05), based on the median across providers serving the model.
Specifications
Technical specifications
| Model name | Llama 4 Scout |
|---|---|
| Provider | Meta |
| Release date | April 5, 2025 |
| Intelligence score | 10 |
| Speed | 108.5 tokens/sec |
| Latency (TTFT) | 0.86s |
| Input price | $0.17 / 1M tokens |
| Output price | $0.66 / 1M tokens |
| Context window | 10,000,000 tokens |
| Parameters | 109B |
| 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-scout |
Pricing
Token prices
Price per 1M tokens
Llama 4 Scout, as listed in the catalog record
- Input$0.17
- Output$0.66
| Token type | Price per 1M |
|---|---|
| Input | $0.17 |
| Output | $0.66 |
Model Intelligence legacy profile: llama-4-scout
FAQ · 22
Frequently asked questions
Llama 4 Scout was released on April 5, 2025.
Llama 4 Scout was created by Meta.
Llama 4 Scout scores 10 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 7).
Llama 4 Scout generates output at 108.5 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s).
Llama 4 Scout has a time to first token (TTFT) of 0.86s (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: 1.59s).
Llama 4 Scout costs $0.17 per 1M input tokens (very competitive, median: $0.53) and $0.66 per 1M output tokens (better than average, median: $1.05), based on the median across providers serving the model.
Llama 4 Scout costs $0.17 per 1M input tokens and $0.66 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 $0.22 per 1M tokens. Pricing may vary by provider.
When evaluated on the Intelligence Index, Llama 4 Scout generated 4.1M output tokens, which is very competitive compared to other open weight non-reasoning models of similar size (median: 9.2M).
No, Llama 4 Scout is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Llama 4 Scout supports text and image input.
Llama 4 Scout supports text output.
Yes, Llama 4 Scout supports image input and can analyze, describe, and answer questions about images.
Yes, Llama 4 Scout is multimodal. It can process text and image input and generate text output.
Llama 4 Scout has a context window of 10M tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Llama 4 Scout is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Llama 4 Scout has 109 billion parameters (17 billion active).
Llama 4 Scout is a Mixture of Experts (MoE) model with 109 billion total parameters, but only 17 billion active parameters are used during inference.
Llama 4 Scout is released under the LLAMA 4 COMMUNITY LICENSE AGREEMENT license. This license allows commercial use.
Llama 4 Scout achieves a score of 10 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Llama 4 Scout has a knowledge cutoff of August 2024. The model's training data includes information up to this date.
Yes, Llama 4 Scout is available via API through 9 providers.
Llama 4 Scout is available through 9 API providers.