MMeta · Released Apr 5, 2025
Llama 4 Scout
Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...
- Released
- Apr 5 2025
- Context window
- 1.31M 16K out
- Input / 1M tokens
- $0.10 from $0.08
- Output / 1M tokens
- $0.30 from $0.24
Where to run it
6 providers compared
Cheapest: DeepInfra, Kilo Gateway at $0.08 / $0.24 per 1M input / output tokens — 67% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| DeepInfraFP8 | $0.08 | $0.24 | — | 328K | 295K | 65.7%24h | Cheapest |
| Kilo Gateway | $0.08 | $0.24 | — | 328K | 16K | — | Cheapest |
| Helicone | $0.08 | $0.30 | — | 131K | 8K | — | +13% |
| DeepInfraFP8 · no tool calling | $0.10 | $0.30 | — | 328K | 16K | 99.5%24h | +25% |
| NanoGPT | $0.085 | $0.46 | $0.043 | 328K | 66K | — | +49% |
| DeepInfrapriority · FP8 | $0.15 | $0.45 | — | 328K | 295K | 95%24h | +87% |
| Vercel AI Gateway | $0.17 | $0.66 | — | 128K | 8K | — | +144% |
| Googleus-east5 | $0.25 | $0.70 | — | 1.31M | 8K | — | +202% |
Benchmarks
Benchmark scores
No benchmark scores yet — Epoch AI hasn't evaluated Llama 4 Scout. Scores appear here automatically once it does.
Specs
Details
- Model ID
meta-llama/llama-4-scout- Developer
- Meta
- Release date
- Apr 5, 2025 ·
- Context window
- 1,310,720 tokens
- Max output
- 16,384 tokens
- API price
- $0.10 / $0.30 per 1M input / output tokens
- Providers
- 6
- Input types
- text, image
- Reasoning
- No
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.