SmolLM3 Instruct: pricing, context window and benchmarks
SmolLM3 Instruct is a open-weight model from Hugging Face. It offers a 8,192-token context window, costs $1.69 per million input tokens and $8.67 per million output tokens, and averages 63.6 across the index's benchmark suite.
Pricing
SmolLM3 Instruct costs $1.69 per million input tokens and $8.67 per million output tokens. A blended 3:1 input-to-output mix works out to about $3.44 per million tokens. Cached input is listed at $0.20 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $1.69 | per 1M tokens |
| Output | $8.67 | per 1M tokens |
| Cached input | $0.20 | per 1M tokens |
| Blended 3:1 | $3.44 | estimate per 1M tokens |
| 1M in + 1M out | $10.36 | 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 | 67.3 | 428 | 29th | |
| GPQA Diamond | 58.4 | 423 | 30th | |
| SWE-bench Verified | 50.6 | 417 | 31th | |
| HumanEval | 69.2 | 408 | 32th | |
| MATH-500 | 67.0 | 409 | 32th | |
| AIME 2025 | 54.4 | 419 | 30th | |
| LiveCodeBench | 56.1 | 397 | 34th | |
| IFEval | 74.0 | 406 | 32th | |
| TruthfulQA | 66.5 | 438 | 27th | |
| HellaSwag | 75.1 | 440 | 27th | |
| MMMU | 60.5 | 425 | 29th | |
| Arena Elo | 1297.8 | 423 | 30th |
What SmolLM3 Instruct is good for
SmolLM3 Instruct 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 $1.69 per million input tokens (hosted list price) and a 8,192-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $1.69 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.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Embed 4 Turbo | Cohere | 4096 | $5.13 | 64.0 |
| Embed 4 | Cohere | 32768 | $4.79 | 63.8 |
| SmolLM3 Instruct | Hugging Face | 8192 | $1.69 | 63.6 |
| Embed 4 Pro | Cohere | 4096 | $5.05 | 63.5 |
| Llama 4 Maverick 8B | Meta AI | 4096 | $1.64 | 63.4 |
| Fusion Mini | OpenRouter | 4096 | $4.89 | 63.3 |
FAQ
How much does SmolLM3 Instruct cost?
SmolLM3 Instruct is listed at $1.69 per million input tokens and $8.67 per million output tokens. Cached input is $0.20 per million tokens. A blended 3:1 input-to-output mix is about $3.44 per million tokens.
What is the context window of SmolLM3 Instruct?
SmolLM3 Instruct has a context window of 8,192 tokens and can emit up to 8,192 output tokens per request in this index.
Is SmolLM3 Instruct open weights?
Yes. SmolLM3 Instruct is listed as an open-weight model under a open weights licence.
How does SmolLM3 Instruct score on benchmarks?
SmolLM3 Instruct averages 63.6 across the scored benchmarks in this index, ranking #412 of 600 models. Its strongest relative result is Arena Elo at 1297.8.
What is SmolLM3 Instruct best for?
SmolLM3 Instruct 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 $1.69 per million input tokens (hosted list price) and a 8,192-token context window, it is a practical option for teams that want more control than a closed API allows.