Qwen3-Coder 8B: pricing, context window and benchmarks
Qwen3-Coder 8B is a coding model from Alibaba Cloud. It offers a 32,768-token context window, costs $1.43 per million input tokens and $7.30 per million output tokens, and averages 61.3 across the index's benchmark suite.
Pricing
Qwen3-Coder 8B costs $1.43 per million input tokens and $7.30 per million output tokens. A blended 3:1 input-to-output mix works out to about $2.90 per million tokens. Cached input is listed at $0.17 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $1.43 | per 1M tokens |
| Output | $7.30 | per 1M tokens |
| Cached input | $0.17 | per 1M tokens |
| Blended 3:1 | $2.90 | estimate per 1M tokens |
| 1M in + 1M out | $8.73 | 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 | 67.6 | 423 | 30th | |
| GPQA Diamond | 55.4 | 459 | 24th | |
| SWE-bench Verified | 46.0 | 453 | 25th | |
| HumanEval | 64.6 | 446 | 26th | |
| MATH-500 | 63.5 | 430 | 28th | |
| AIME 2025 | 54.4 | 418 | 30th | |
| LiveCodeBench | 52.4 | 423 | 30th | |
| IFEval | 70.1 | 461 | 23th | |
| TruthfulQA | 64.9 | 461 | 23th | |
| HellaSwag | 76.6 | 419 | 30th | |
| MMMU | 58.8 | 444 | 26th | |
| Arena Elo | 1286.5 | 436 | 27th |
What Qwen3-Coder 8B is good for
Qwen3-Coder 8B is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $1.43 per million input tokens, it sits in the mid-range of the index, and the 32,768-token context window is enough for large files and multi-file diffs.
Strengths
- Input price of $1.43 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- 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 |
|---|---|---|---|---|
| Firesearch Mini | Fireworks AI | 32768 | $3.94 | 61.5 |
| 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 |
FAQ
How much does Qwen3-Coder 8B cost?
Qwen3-Coder 8B is listed at $1.43 per million input tokens and $7.30 per million output tokens. Cached input is $0.17 per million tokens. A blended 3:1 input-to-output mix is about $2.90 per million tokens.
What is the context window of Qwen3-Coder 8B?
Qwen3-Coder 8B has a context window of 32,768 tokens and can emit up to 32,768 output tokens per request in this index.
Is Qwen3-Coder 8B open weights?
Yes. Qwen3-Coder 8B is listed as an open-weight model under a open weights licence.
How does Qwen3-Coder 8B score on benchmarks?
Qwen3-Coder 8B averages 61.3 across the scored benchmarks in this index, ranking #440 of 600 models. Its strongest relative result is Arena Elo at 1286.5.
What is Qwen3-Coder 8B best for?
Qwen3-Coder 8B is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $1.43 per million input tokens, it sits in the mid-range of the index, and the 32,768-token context window is enough for large files and multi-file diffs.