Llama 3 Anyscale 1.5B: pricing, context window and benchmarks
Llama 3 Anyscale 1.5B is a small open-weight model from Anyscale. It offers a 16,384-token context window, costs $0.43 per million input tokens and $2.27 per million output tokens, and averages 48.9 across the index's benchmark suite.
Pricing
Llama 3 Anyscale 1.5B costs $0.43 per million input tokens and $2.27 per million output tokens. A blended 3:1 input-to-output mix works out to about $0.89 per million tokens. Cached input is listed at $0.04 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $0.43 | per 1M tokens |
| Output | $2.27 | per 1M tokens |
| Cached input | $0.04 | per 1M tokens |
| Blended 3:1 | $0.89 | estimate per 1M tokens |
| 1M in + 1M out | $2.70 | 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 | 56.2 | 567 | 6th | |
| GPQA Diamond | 42.7 | 577 | 4th | |
| SWE-bench Verified | 35.7 | 549 | 9th | |
| HumanEval | 50.5 | 570 | 5th | |
| MATH-500 | 47.4 | 572 | 5th | |
| AIME 2025 | 35.1 | 569 | 5th | |
| LiveCodeBench | 35.8 | 563 | 6th | |
| IFEval | 62.3 | 553 | 8th | |
| TruthfulQA | 56.8 | 554 | 8th | |
| HellaSwag | 67.0 | 562 | 6th | |
| MMMU | 48.8 | 557 | 7th | |
| Arena Elo | 1175.9 | 579 | 4th |
What Llama 3 Anyscale 1.5B is good for
Llama 3 Anyscale 1.5B 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 $0.43 per million input tokens it can handle a large share of production traffic.
Strengths
- Input price of $0.43 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 16,384 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 |
|---|---|---|---|---|
| Llama 3 Replicate 1.5B | Replicate | 4096 | $0.51 | 49.3 |
| Mistral Small 3 1.5B | Mistral AI | 32768 | $0.42 | 49.2 |
| Llama 3 Novita 1.5B | Novita AI | 4096 | $0.41 | 49.1 |
| Gemma 3 1.5B | Google DeepMind | 32768 | $0.40 | 48.9 |
| Llama 3 Anyscale 1.5B | Anyscale | 16384 | $0.43 | 48.9 |
| Llama 3 DeepInfra 1.5B | DeepInfra | 32768 | $0.48 | 48.8 |
FAQ
How much does Llama 3 Anyscale 1.5B cost?
Llama 3 Anyscale 1.5B is listed at $0.43 per million input tokens and $2.27 per million output tokens. Cached input is $0.041 per million tokens. A blended 3:1 input-to-output mix is about $0.89 per million tokens.
What is the context window of Llama 3 Anyscale 1.5B?
Llama 3 Anyscale 1.5B has a context window of 16,384 tokens and can emit up to 16,384 output tokens per request in this index.
Is Llama 3 Anyscale 1.5B open weights?
Yes. Llama 3 Anyscale 1.5B is listed as an open-weight model under a open weights licence.
How does Llama 3 Anyscale 1.5B score on benchmarks?
Llama 3 Anyscale 1.5B averages 48.9 across the scored benchmarks in this index, ranking #568 of 600 models. Its strongest relative result is Arena Elo at 1175.9.
What is Llama 3 Anyscale 1.5B best for?
Llama 3 Anyscale 1.5B 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 $0.43 per million input tokens it can handle a large share of production traffic.