What can a RTX 3080 10GB run?
10GB · 760 GB/s · 320W · $420
Fast, hot, and 10 GB — a sprightly racehorse for 7–8B models when the price is right.
Runs great14
Comfortable headroom — daily-driver territory.
| Model | Quant | Est. speed | Max context |
|---|---|---|---|
| deepseek moe 16b chat | FP16 | Infinity tok/s | — |
| qwen2 7b instruct | FP16 | Infinity tok/s | — |
| Qwen3-0.6B | FP16 | 379 tok/s | 33k |
| Gemma-3n-E4B | Q8_0 | 316 tok/s | 49k |
| SmolLM2-1.7B | FP16 | 138 tok/s | 8k |
| Llama-3.2-3B | Q8_0 | 138 tok/s | 60k |
| Phi-4-mini | Q8_0 | 115 tok/s | 50k |
| Gemma-3-4B | Q8_0 | 115 tok/s | 49k |
| Qwen3-4B | Q8_0 | 113 tok/s | 33k |
| Llama-3.1-8B | Q4_K_M | 97 tok/s | 32k |
| Qwen3-8B | Q4_K_M | 96 tok/s | 32k |
| Mistral-7B-Instruct | Q5_K_M | 91 tok/s | 31k |
| DeepSeek-R1-Distill-Qwen-7B | Q5_K_M | 87 tok/s | 27k |
| Qwen2.5-Coder-7B | Q5_K_M | 87 tok/s | 27k |
Tight squeeze6
It fits, but close the browser tabs and mind the context.
| Model | Quant | Est. speed | Max context |
|---|---|---|---|
| GPT-OSS-20B | IQ2_M | 387 tok/s | 10k |
| Qwen3-14B | IQ2_M | 83 tok/s | 21k |
| Phi-4-14B | Q3_K_M | 66 tok/s | 10k |
| DeepSeek-R1-Distill-Qwen-14B | Q3_K_M | 65 tok/s | 10k |
| Codestral-22B | IQ2_M | 63 tok/s | 7k |
| Gemma-3-12B | Q4_K_M | 60 tok/s | 5k |
Measured on this card0
Everything here was benchmarked on this exact card by someone who owns one — the tables above are our arithmetic; these numbers came off real machines.
Nobody has measured anything on this card yet. Paste a benchmark — it takes one keystroke and no form.