Reference

Models that fit in 8 GB of VRAM

A 8 GB card has about 7.4 GB usable once the driver and desktop take their share. Every model below is sized against that budget, including the KV cache for a working context.

26 models fit7.4 GB usableEstimated

Top picks

Best on 8 GB VRAM

Best overallGood fit

Qwen2.5 Coder 7B Instruct

The practical local copilot. Supports fill-in-the-middle, so it works as an inline completion model rather than only a chat assistant.

Params: 7.6B
Quant: Q4_K_M
Memory: 5.3 GB

Runner-upGood fit

Qwen2.5 7B Instruct

The default starting point for local work on 8–12 GB cards. Strong instruction following and reliable tool-call formatting for its size.

Params: 7.6B
Quant: Q4_K_M
Memory: 5.3 GB

ThirdGood fit

Mistral 7B Instruct v0.3

The model that made local AI practical. Still a dependable, fast, unconditionally Apache-licensed baseline.

Params: 7.25B
Quant: Q4_K_M
Memory: 5.7 GB

Full results

26 models

Estimated

#	Model	Params	Licence	Quantization	Memory	Fit	Fine-tune
1	Qwen2.5 Coder 7B Instruct	7.6B	Apache 2.0	Q4_K_M	5.3 GB	Good fit	Yes
2	Qwen2.5 7B Instruct	7.6B	Apache 2.0	Q4_K_M	5.3 GB	Good fit	Yes
3	Mistral 7B Instruct v0.3	7.25B	Apache 2.0	Q4_K_M	5.7 GB	Good fit	Yes
4	Gemma 3 4B Instruct	4.3B	Gemma Terms of Use	Q4_K_M	4.3 GB	Excellent fit	Yes
5	Phi-3.5 Mini Instruct	3.8B	MIT	Q4_K_M	5.7 GB	Good fit	Yes
6	Llama 3.1 8B Instruct	8B	Llama 3.1 Community License	Q4_K_M	6.1 GB	Tight fit	Yes
7	Llama 3.2 3B Instruct	3.2B	Llama 3.2 Community License	Q4_K_M	3.2 GB	Excellent fit	Yes
8	Qwen3 8B	8.2B	Apache 2.0	Q4_K_M	6.3 GB	Tight fit	Yes
9	Qwen2.5 Coder 32B Instruct	32.8B	Apache 2.0	Q3_K_M	18 GB	Runs with CPU offload	No
10	Qwen2.5 32B Instruct	32.8B	Apache 2.0	Q3_K_M	18 GB	Runs with CPU offload	No
11	DeepSeek-R1-Distill-Qwen-32B	32.8B	MIT	Q3_K_M	18 GB	Runs with CPU offload	No
12	Gemma 3 27B Instruct	27.4B	Gemma Terms of Use	Q3_K_M	19 GB	Runs with CPU offload	No
13	Phi-4 14B	14.7B	MIT	Q4_K_M	11 GB	Runs with CPU offload	No
14	Mistral Small 24B Instruct	23.6B	Apache 2.0	Q4_K_M	16 GB	Runs with CPU offload	No
15	SmolLM2 1.7B Instruct	1.7B	Apache 2.0	F16	5.2 GB	Good fit	Yes
16	Qwen3 30B-A3B	30.5B	Apache 2.0	Q4_K_M	19 GB	Runs with CPU offload	No
17	Qwen2.5 14B Instruct	14.8B	Apache 2.0	Q4_K_M	11 GB	Runs with CPU offload	No
18	Llama 3.2 1B Instruct	1.24B	Llama 3.2 Community License	F16	3.0 GB	Excellent fit	Yes
19	StarCoder2 15B	16B	BigCode OpenRAIL-M	Q4_K_M	10 GB	Runs with CPU offload	No
20	Qwen3 14B	14.8B	Apache 2.0	Q4_K_M	10 GB	Runs with CPU offload	No
21	DeepSeek-R1-Distill-Qwen-14B	14.8B	MIT	Q4_K_M	11 GB	Runs with CPU offload	No
22	Mistral Nemo 12B Instruct	12.2B	Apache 2.0	Q3_K_M	7.8 GB	Runs with CPU offload	No
23	Gemma 3 12B Instruct	12.2B	Gemma Terms of Use	Q4_K_M	10 GB	Runs with CPU offload	No
24	DeepSeek-Coder-V2-Lite Instruct	15.7B	DeepSeek License	Q4_K_M	11 GB	Runs with CPU offload	No
25	Gemma 2 9B Instruct	9.2B	Gemma Terms of Use	Q4_K_M	8.1 GB	Runs with CPU offload	No
26	Code Llama 7B Instruct	6.7B	Llama 2 Community License	Q5_K_M	9.0 GB	Runs with CPU offload	Yes

How this page is produced

Each model is sized at every quantization it publishes: weights at the true bits-per-weight, KV cache for the context window, plus runtime overhead. The recommendation is the highest quality that still leaves room for a real context.

· Token rates are omitted here: a capacity is not a specific card, and generation speed depends on memory bandwidth rather than size.
· For a real machine, use a hardware page or detect your own in the workspace.
· 11 of these can also be fine-tuned on this hardware.

Check your own machine

ModelLM can read your actual GPU, VRAM and RAM and size every model against it.

Detect my hardware

Models that fit in 6 GB of VRAM Models that fit in 12 GB of VRAM Models that fit in 16 GB of VRAM Models that fit in 24 GB of VRAM Models that fit in 32 GB of VRAM Models that fit in 48 GB of VRAM