# Models

Every plan includes the launch model with no token meter. Models marked coming soon are not available at launch. Source data: [models.json](https://kilowatt.sh/models.json). Lineup updated 2026-11-10; it may change as better open-weight coding models ship.

| Model | Id | Availability | Maker | Weights / license | Context |
| --- | --- | --- | --- | --- | --- |
| [DeepSeek V4.1 Flash-1017](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) | `deepseek-v4.1-flash` | At launch | DeepSeek | Open weights, license: [MIT](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) | see model card |
| [GLM-5.3-Flash](https://huggingface.co/zai-org/GLM-5.3-Flash) | `glm-5.3-flash` | **Coming soon** (not at launch) | Z.ai | Open weights, license: [MIT](https://huggingface.co/zai-org/GLM-5.3-Flash) | see model card |
| Router | `auto` | At launch | Kilowatt | n/a | n/a |

### DeepSeek V4.1 Flash-1017

- Id: `deepseek-v4.1-flash`
- Availability: at launch
- Maker: DeepSeek
- Size: 552B backbone MoE + 196B Engram memory (763B total per Hugging Face); 8B active in prefill, 16B in decode
- License: [MIT](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017)
- Router role: Default model at launch; serves every request until a second model is added.
- Model card: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017

### GLM-5.3-Flash (coming soon)

- Id: `glm-5.3-flash`
- Availability: coming soon, not at launch
- Maker: Z.ai
- Size: 320B total / 18B active; hybrid sparse + linear attention; multimodal
- License: [MIT](https://huggingface.co/zai-org/GLM-5.3-Flash)
- Router role: Added to the auto router for tool-heavy and vision requests when it ships.
- Model card: https://huggingface.co/zai-org/GLM-5.3-Flash

## Router

Model id `auto`: At launch it sends every request to DeepSeek V4.1 Flash-1017. When GLM-5.3-Flash is added, auto will pick a model per request; you can always pin a model id instead.

## Vendor-reported scores

We haven't published our own benchmarks yet. These are the model makers' self-reported scores; each number links to its source. Harnesses differ, so this is not a head-to-head comparison, and results through any provider can differ.

| Benchmark (vendor-reported) | DeepSeek V4.1 Flash-1017 | GLM-5.3-Flash (coming soon) |
| --- | --- | --- |
| DeepSWE v1.1 (resolved) | [77.0](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) (self-reported) | [63.4](https://z.ai/blog/glm-5.3-flash) (self-reported) |
| Terminal-Bench 4.0 (pass@1) | [42.3](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) (self-reported) | — |
| Terminal-Bench 2.1 (pass@1) | — | [84.3](https://z.ai/blog/glm-5.3-flash) (self-reported) |
| Toolathlon Verified (pass@1) | — | [78.4](https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/.eval_results/GLM-5.3-Flash.yaml) (self-reported) |

- DeepSeek V4.1 Flash-1017: [model card](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017). Self-reported by DeepSeek on the model card.
- GLM-5.3-Flash: [model card](https://huggingface.co/zai-org/GLM-5.3-Flash). Self-reported by Z.ai (blog and Hugging Face eval results).

---
Source: https://kilowatt.sh/docs/models · Markdown: https://kilowatt.sh/docs/models.md · All docs: https://kilowatt.sh/llms-full.txt