Is 10GB of VRAM enough to run AI models locally?

Yes — 156 of the 228 models tested run entirely on the GPU with 10GB at 4K context, and 44 more run with part of the model offloaded to system RAM. The heaviest that fits is nous-hermes2:10.7b-solar-q4_K_M, needing about 9.0 GB.

What is the best AI model for 10GB VRAM?

For maximum capability, nous-hermes2 (nous-hermes2:10.7b-solar-q4_K_M) is the heaviest build that still fits entirely in 10GB. Smaller models leave headroom for a longer context window, which often matters more than raw parameter count.

Which graphics cards have 10GB of VRAM?

Cards with 10GB include GeForce RTX 3080.

What AI models can you run with 10GB of VRAM?

With 10GB of VRAM, 156 of 228 models from the Ollama library run entirely on the GPU at 4K context. 44 more run by offloading part of the model to system RAM, which works but is slower.

156

run well

with trade-offs

228

models tested

Check my actual PC

Best AI models for 10GB of VRAM

Heaviest first — each of these fits entirely in 10GB at 4K context, counting weights, KV cache, compute buffers and runtime overhead.

Models that fit in 10GB of VRAM
Model	Build	Memory needed	Verdict
nous-hermes2	10.7B	~9.0 GB	Good fit
solar	10.7B	~9.0 GB	Good fit
codegeex4	9B	~8.7 GB	Good fit
glm4	9B	~8.7 GB	Good fit
ornith-1.5	9B	~7.9 GB	Good fit
ornith	9B	~7.9 GB	Good fit
gemma3n	e2b-it-q4_K_M	~7.9 GB	Good fit
lfm2.5	8B	~7.3 GB	Excellent fit
granite4.1-guardian	8B	~7.2 GB	Excellent fit
rnj-1	8B	~7.2 GB	Excellent fit
command-r7b-arabic	7B	~7.1 GB	Excellent fit
command-r7b	7B	~7.1 GB	Excellent fit
aya-expanse	8B	~7.1 GB	Excellent fit
llama-pro	8B	~7.1 GB	Excellent fit
minicpm-v4.5	8B	~7.1 GB	Excellent fit
llava-llama3	8B	~6.9 GB	Excellent fit
tulu3	8B	~6.9 GB	Excellent fit
llama3-groq-tool-use	8B	~6.9 GB	Excellent fit
dolphin3	8B	~6.9 GB	Excellent fit
dolphin-llama3	8B	~6.9 GB	Excellent fit
llama3.1	8B	~6.9 GB	Excellent fit
llama3-gradient	8B	~6.9 GB	Excellent fit
llama3	8B	~6.9 GB	Excellent fit
llama3-chatqa	8B	~6.9 GB	Excellent fit
aya	8B	~6.8 GB	Excellent fit

Graphics cards with 10GB of VRAM

GeForce RTX 3080

How this was calculated

Every model is evaluated at 4K context against 10GB of VRAM, minus a reserve for the display and operating system, and assuming 32GB of system RAM with working GPU drivers. Your own machine may differ, which is what the PC scan measures directly.

10GB VRAM and local AI — frequently asked questions

Is 10GB of VRAM enough to run AI models locally?: Yes — 156 of the 228 models tested run entirely on the GPU with 10GB at 4K context, and 44 more run with part of the model offloaded to system RAM. The heaviest that fits is nous-hermes2:10.7b-solar-q4_K_M, needing about 9.0 GB.
What is the best AI model for 10GB VRAM?: For maximum capability, nous-hermes2 (nous-hermes2:10.7b-solar-q4_K_M) is the heaviest build that still fits entirely in 10GB. Smaller models leave headroom for a longer context window, which often matters more than raw parameter count.
Which graphics cards have 10GB of VRAM?: Cards with 10GB include GeForce RTX 3080.

Other VRAM amounts

4GB 6GB 8GB 12GB 16GB 24GB 32GB