What AI models run on the RTX 3080 10GB?
The RTX 3080 10GB has 10 GB of VRAM (760 GB/s). About 9.5 GB is usable for a model. 165 of 298 open models we track fit at 8K context.
This card has 10 GB of VRAM. See every model that fits in 12 GB VRAM.
Compare for local LLMs: RTX 3080 10GB vs RTX 4070 all comparisons
Using a specific tool? Ollama models that run on the RTX 3080 10GB · LM Studio · llama.cpp offload
Models that fit, largest first
Bigger models with CPU offload
These popular models don't fit in the RTX 3080 10GB's VRAM at 4-bit, but run with llama.cpp offload, keeping part of the model in system RAM. Speed range is DDR4-3200 to DDR5-6000 (estimate). Method.
| Model | Params | Offload plan |
|---|---|---|
| GLM 5.3 Flash | 321B MoE | Offload Q2_K, 128 GB RAM · ~6.9–14 tok/s |
| Gemma 4 31B IT | 31B | Offload Q4_K_M, 32 GB RAM · ~1.9–3.8 tok/s |
| gpt-oss 120B | 117B MoE | Offload Q4_K_M, 64 GB RAM · ~13–26 tok/s |
| DeepSeek V4 Flash 0731 | 304B MoE | Offload Q2_K, 128 GB RAM · ~7.9–15 tok/s |
| DeepSeek V4 Flash | 291B MoE | Offload Q2_K, 128 GB RAM · ~9.2–18 tok/s |
| GLM 4.5 Air | 110B MoE | Offload Q4_K_M, 128 GB RAM · ~5.2–10 tok/s |
| MiniMax M2.5 | 229B MoE | Offload Q3_K_M, 128 GB RAM · ~6.7–14 tok/s |
| Hy3 | 299B MoE | Offload Q2_K, 128 GB RAM · ~5.3–11 tok/s |
| Nex N2.6 | 321B MoE | Offload Q2_K, 128 GB RAM · ~6.6–13 tok/s |
Comparing cards? The RTX 3080 10GB (launch MSRP $699) is in best GPUs for local LLMs under $800. See also Mac vs PC.
Just too big at 4-bit
These need up to 35% more memory than the RTX 3080 10GB has at Q4_K_M and 8K context. A smaller quant or shorter context may work.
| Model | Q4_K_M needs | Short by |
|---|---|---|
| DeepSeek V2 Lite | 9.8 GB | 0.3 GB |
| LLaDA2.1 Mini | 10.3 GB | 0.8 GB |
| Kimi VL A3B Thinking | 10.3 GB | 0.8 GB |
| Qwen3 14B | 10.3 GB | 0.8 GB |
| DeepSeek Coder V2 Lite Instruct | 10.5 GB | 1.0 GB |
| DeepSeek V2 Lite Chat | 10.5 GB | 1.0 GB |
| DeepSeek R1 Distill Qwen 14B | 10.6 GB | 1.1 GB |
| Qwen2.5 Coder 14B Instruct | 10.6 GB | 1.1 GB |
| Phi 4 | 10.6 GB | 1.1 GB |
| Moonlight 16B A3B Instruct | 10.8 GB | 1.3 GB |
| Kimi VL A3B Thinking 2506 | 10.8 GB | 1.3 GB |
| Kimi VL A3B Instruct | 10.8 GB | 1.3 GB |
| gpt-oss Safeguard 20B | 11.8 GB | 2.3 GB |
| gpt-oss 20B | 11.8 GB | 2.3 GB |
| DeepSeek MoE 16B Base | 12.5 GB | 3.0 GB |
Questions
What is the biggest model that runs on the RTX 3080 10GB?
Of the models we track, the largest that fits at 8K context is LFM2.5 24B A2B (24B) at Q2_K, needing about 9.2 GB.
How much of the RTX 3080 10GB's memory can a model use?
About 9.5 GB: we assume 95% of the 10 GB is usable after driver and display overhead.
Spec source: NVIDIA specifications. Model data: Hugging Face Hub API. Speed is a bandwidth ceiling (bandwidth ÷ bytes read per token), not a benchmark. Methodology · Report an error.