Configure a rig →

Solar10.7B params4k context

Solar 10.7B locally

Solar 10.7B is an open-weight LLM from the Solar family with 10.7B parameters. Main use: chat, RAG and general assistance. Detected minimum hardware: GTX 1070 (8 GB).

Check on my GPU →Hugging Face ↗

Technical facts

Parameters10.7B

Max context4k

Q4_K_M6.7 GB

Q5_K_M8.2 GB

Q812.0 GB

FP1623.9 GB

FamilySolar

Last sync2026-05-12

Available quantizations

GGUF weights

Q4_K_M

6.7GB

Acceptable. Good compromise when VRAM is limited.

Q5_K_M

8.2GB

Good quality. Sweet spot for size and precision.

Q8

12.0GB

Near-FP16 quality. Comfortable for production.

FP16

23.9GB

Reference precision. Maximum quality, doubled VRAM.

Compatible GPUs

12 single-GPU

GPUs that can run Solar 10.7B on a single card, ranked by VRAM headroom.

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

6.7 / 8 GBcomfortable · Q4

Recommended multi-GPU rigs

2x / 4x consumer GPUs

For Solar 10.7B at higher quantization or with more context, a multi-GPU rig gives more headroom.

6.7 / 8 GBcomfortable · Q4

2× GTX 1060 6GB

12 GB · GTX 10

8.2 / 12 GBcomfortable · Q5

12 GB · GTX 16

8.2 / 12 GBcomfortable · Q5

2× GTX 1660 Super

12 GB · GTX 16

8.2 / 12 GBcomfortable · Q5

2× GTX 1660 Ti

12 GB · GTX 16

8.2 / 12 GBcomfortable · Q5

2× RTX 2060 6GB

12 GB · RTX 20

8.2 / 12 GBcomfortable · Q5

2× RTX 3050 6GB

12 GB · RTX 30

8.2 / 12 GBcomfortable · Q5

12 GB · Arc Alchemist

8.2 / 12 GBcomfortable · Q5

Recommended rig

2× GTX 1650

Solar 10.7B with Ubuntu, vLLM, Open WebUI and the model already downloaded.

VRAM estimates: parameters x bits/8 plus margin. Real performance varies by engine, context length and batch size.
sync: 2026-05-12