Llama 3 Anyscale Base: pricing, context window and benchmarks
Llama 3 Anyscale Base is a open-weight model from Anyscale. It offers a 65,536-token context window, costs $3.23 per million input tokens and $14.41 per million output tokens, and averages 72.9 across the index's benchmark suite.
Pricing
Llama 3 Anyscale Base costs $3.23 per million input tokens and $14.41 per million output tokens. A blended 3:1 input-to-output mix works out to about $6.03 per million tokens. Cached input is listed at $0.38 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.23 | per 1M tokens |
| Output | $14.41 | per 1M tokens |
| Cached input | $0.38 | per 1M tokens |
| Blended 3:1 | $6.03 | estimate per 1M tokens |
| 1M in + 1M out | $17.64 | 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 | 74.7 | 340 | 43th | |
| GPQA Diamond | 69.9 | 267 | 56th | |
| SWE-bench Verified | 58.3 | 350 | 42th | |
| HumanEval | 75.6 | 333 | 45th | |
| MATH-500 | 77.0 | 276 | 54th | |
| AIME 2025 | 69.2 | 275 | 54th | |
| LiveCodeBench | 63.5 | 331 | 45th | |
| IFEval | 81.6 | 283 | 53th | |
| TruthfulQA | 77.8 | 258 | 57th | |
| HellaSwag | 83.7 | 291 | 52th | |
| MMMU | 70.4 | 284 | 53th | |
| Arena Elo | 1379.8 | 273 | 55th |
What Llama 3 Anyscale Base is good for
Llama 3 Anyscale Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.23 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.23 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Claude 3.7 Sonnet Mini | Anthropic | 32768 | $9.74 | 73.0 |
| Nemotron 4 Base | NVIDIA | 8192 | $3.29 | 73.0 |
| Firefunction Pro | Fireworks AI | 16384 | $9.05 | 72.9 |
| Llama 3 Anyscale Base | Anyscale | 65536 | $3.23 | 72.9 |
| Llama 3 Lambda Base | Lambda | 8192 | $3.19 | 72.9 |
| Mixtral Instruct | Mistral AI | 65536 | $3.35 | 72.9 |
FAQ
How much does Llama 3 Anyscale Base cost?
Llama 3 Anyscale Base is listed at $3.23 per million input tokens and $14.41 per million output tokens. Cached input is $0.38 per million tokens. A blended 3:1 input-to-output mix is about $6.03 per million tokens.
What is the context window of Llama 3 Anyscale Base?
Llama 3 Anyscale Base has a context window of 65,536 tokens and can emit up to 16,384 output tokens per request in this index.
Is Llama 3 Anyscale Base open weights?
Yes. Llama 3 Anyscale Base is listed as an open-weight model under a open weights licence.
How does Llama 3 Anyscale Base score on benchmarks?
Llama 3 Anyscale Base averages 72.9 across the scored benchmarks in this index, ranking #305 of 600 models. Its strongest relative result is Arena Elo at 1379.8.
What is Llama 3 Anyscale Base best for?
Llama 3 Anyscale Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.23 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.