Llama 4 Scout 8B: pricing, context window and benchmarks
Llama 4 Scout 8B is a small open-weight model from Meta AI. It offers a 8,192-token context window, costs $1.46 per million input tokens and $6.51 per million output tokens, and averages 61.2 across the index's benchmark suite.
Pricing
Llama 4 Scout 8B costs $1.46 per million input tokens and $6.51 per million output tokens. A blended 3:1 input-to-output mix works out to about $2.72 per million tokens. Cached input is listed at $0.14 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $1.46 | per 1M tokens |
| Output | $6.51 | per 1M tokens |
| Cached input | $0.14 | per 1M tokens |
| Blended 3:1 | $2.72 | estimate per 1M tokens |
| 1M in + 1M out | $7.97 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 67.5 | 425 | 29th | |
| GPQA Diamond | 58.9 | 418 | 30th | |
| SWE-bench Verified | 48.1 | 436 | 27th | |
| HumanEval | 66.3 | 426 | 29th | |
| MATH-500 | 62.7 | 439 | 27th | |
| AIME 2025 | 50.2 | 458 | 24th | |
| LiveCodeBench | 49.7 | 452 | 25th | |
| IFEval | 71.8 | 430 | 28th | |
| TruthfulQA | 65.8 | 446 | 26th | |
| HellaSwag | 73.7 | 462 | 23th | |
| MMMU | 58.1 | 457 | 24th | |
| Arena Elo | 1266.8 | 466 | 22th |
What Llama 4 Scout 8B is good for
Llama 4 Scout 8B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $1.46 per million input tokens it can handle a large share of production traffic.
Strengths
- Input price of $1.46 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
Trade-offs
- Context window of 8,192 tokens is small for long-document work.
- Average benchmark score is below the index median; test it on your own task.
- Less reliable on open-ended reasoning and complex multi-step instructions.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| GLM-4.5 8B | Zhipu AI | 32768 | $1.46 | 61.3 |
| Qwen3-Coder 8B | Alibaba Cloud | 32768 | $1.43 | 61.3 |
| Zephyr Chat | Hugging Face | 4096 | $1.49 | 61.3 |
| Llama 4 Scout 8B | Meta AI | 8192 | $1.46 | 61.2 |
| Mistral Large 3 7B | Mistral AI | 32768 | $1.44 | 61.1 |
| Reka Flash Nano | Reka AI | 16384 | $3.93 | 61.0 |
FAQ
How much does Llama 4 Scout 8B cost?
Llama 4 Scout 8B is listed at $1.46 per million input tokens and $6.51 per million output tokens. Cached input is $0.14 per million tokens. A blended 3:1 input-to-output mix is about $2.72 per million tokens.
What is the context window of Llama 4 Scout 8B?
Llama 4 Scout 8B has a context window of 8,192 tokens and can emit up to 8,192 output tokens per request in this index.
Is Llama 4 Scout 8B open weights?
Yes. Llama 4 Scout 8B is listed as an open-weight model under a open weights licence.
How does Llama 4 Scout 8B score on benchmarks?
Llama 4 Scout 8B averages 61.2 across the scored benchmarks in this index, ranking #442 of 600 models. Its strongest relative result is Arena Elo at 1266.8.
What is Llama 4 Scout 8B best for?
Llama 4 Scout 8B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $1.46 per million input tokens it can handle a large share of production traffic.