Model library

# Models

The open models Gridline fine-tunes and runs in batch today, and the ones being qualified next.

## Live today

Each card says what its model is live for: fine-tuning, batch or both, through the OpenAI-compatible API. Limits and settings for each are in the [docs](https://gridlinegpu.com/docs/models).

- ### `qwen3.8-27b`

Apache-2.0

Batch
- ### `nemotron-3.5-lightning-30b-a3b`

OpenMDW-1.1

Batch
- ### `gemma4-26b-a4b-it`

Apache-2.0

Batch
- ### `qwen3.5-9b`

Apache-2.0

Fine-tuning, batch
- ### `gpt-oss-20b`

Apache-2.0

Fine-tuning, batch
- ### `qwen3-embedding-0.6b`

Apache-2.0

Batch embeddings
- ### `qwen3-8b`

Apache-2.0

Fine-tuning, batch
- ### `qwen2.5-0.5b-instruct`

Apache-2.0

Fine-tuning, batch
- ### `qwen2.5-1.5b-instruct`

Apache-2.0

Fine-tuning, batch
- ### `qwen2.5-7b-instruct`

Apache-2.0

Fine-tuning, batch
- ### `llama-3.1-8b-instruct`

Llama 3.1 Community License

Fine-tuning, batch
- ### `mistral-7b-instruct-v0.3`

Apache-2.0

Fine-tuning, batch
- ### `bge-m3`

MIT

Batch embeddings

## In qualification

Being qualified next. Not available yet, and no date is promised.

- ### `gpt-oss-120b`

Apache-2.0

Chat

Gridline is an independent service. Qwen, Mistral, Llama, gpt-oss, Nemotron, Gemma and bge-m3 are open models from their respective developers; Gridline is not affiliated with or endorsed by them. The logos are their owners' trademarks, shown to name who made each model.

---

Source: https://gridlinegpu.com/models