Yi-Lightning: pricing, context window and benchmarks
Yi-Lightning is a multimodal model from 01.AI. It offers a 32,768-token context window, costs $4.22 per million input tokens and $23.07 per million output tokens, and averages 77.2 across the index's benchmark suite.
Pricing
Yi-Lightning costs $4.22 per million input tokens and $23.07 per million output tokens. A blended 3:1 input-to-output mix works out to about $8.93 per million tokens. Cached input is listed at $0.58 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $4.22 | per 1M tokens |
| Output | $23.07 | per 1M tokens |
| Cached input | $0.58 | per 1M tokens |
| Blended 3:1 | $8.93 | estimate per 1M tokens |
| 1M in + 1M out | $27.29 | 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 | 81.2 | 144 | 76th | |
| GPQA Diamond | 71.0 | 239 | 60th | |
| SWE-bench Verified | 65.7 | 200 | 67th | |
| HumanEval | 81.4 | 203 | 66th | |
| MATH-500 | 80.1 | 205 | 66th | |
| AIME 2025 | 70.6 | 246 | 59th | |
| LiveCodeBench | 71.6 | 159 | 74th | |
| IFEval | 85.8 | 166 | 72th | |
| TruthfulQA | 77.8 | 261 | 57th | |
| HellaSwag | 87.5 | 160 | 73th | |
| MMMU | 76.1 | 138 | 77th | |
| Arena Elo | 1396.4 | 221 | 63th |
What Yi-Lightning is good for
Yi-Lightning 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 $4.22 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
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- No standout weakness in the indexed specifications, but real-world behaviour still needs testing.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Kimi K1.5 Base | Moonshot AI | 65536 | $4.34 | 77.4 |
| Qwen3 Base | Alibaba Cloud | 65536 | $4.33 | 77.4 |
| Reka Core | Reka AI | 32768 | $12.17 | 77.4 |
| DeepSeek-Coder Chat | DeepSeek | 200000 | $4.20 | 77.3 |
| Yi-Lightning | 01.AI | 32768 | $4.22 | 77.2 |
| Sonar Pro | Perplexity | 65536 | $11.25 | 77.1 |
FAQ
How much does Yi-Lightning cost?
Yi-Lightning is listed at $4.22 per million input tokens and $23.07 per million output tokens. Cached input is $0.58 per million tokens. A blended 3:1 input-to-output mix is about $8.93 per million tokens.
What is the context window of Yi-Lightning?
Yi-Lightning has a context window of 32,768 tokens and can emit up to 4,096 output tokens per request in this index.
Is Yi-Lightning open weights?
Yes. Yi-Lightning is listed as an open-weight model under a open weights licence.
How does Yi-Lightning score on benchmarks?
Yi-Lightning averages 77.2 across the scored benchmarks in this index, ranking #192 of 600 models. Its strongest relative result is Arena Elo at 1396.4.
What is Yi-Lightning best for?
Yi-Lightning 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 $4.22 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.