Discrete GPU
GeForce RTX 3080
Data as of
- Source
- Checked:
With 10 GB
One entry per model, up to 30, sorted by popularity. For each model we show the largest (most precise) quantization that fits in this memory with a context of at least 8K tokens. The context length is the longest one that quantization fits (for models whose maximum is under 8K, such as embedding models, that maximum). Models that are not for chat or text generation are labelled with their use. "Left" is the memory minus the estimate; under 5% of the memory is marked "tight" (not a guarantee that it fits).
-
Context length 8,192 Estimated 10.7 GB 0.02 GB left (tight)
-
Context length 32,768 Estimated 10.7 GB 0.01 GB left (tight)
-
Context length 8,192 Estimated 10.5 GB 0.22 GB left (tight)
-
Context length 8,192 Estimated 10.7 GB 0.01 GB left (tight)
-
Ornith-1.0-9B Q6_KContext length 65,536 Estimated 10.6 GB 0.10 GB left (tight)
-
Context length 512 Estimated 2.41 GB 8.33 GB left
-
Context length 16,384 Estimated 10.4 GB 0.34 GB left (tight)
-
MOSS-Music-8B-Instruct Q4_K_MContext length 16,384 Estimated 8.45 GB 2.29 GB left
-
Context length 8,192 Estimated 9.55 GB 1.19 GB left
-
Context length 16,384 Estimated 10.4 GB 0.35 GB left (tight)
-
Context length 65,536 Estimated 10.6 GB 0.13 GB left (tight)
-
Ternary-Bonsai-2-27B Q2_0Context length 16,384 Estimated 9.90 GB 0.84 GB left
-
Context length 8,192 Estimated 10.5 GB 0.23 GB left (tight)
-
Context length 16,384 Estimated 10.3 GB 0.44 GB left (tight)
-
Context length 8,192 Estimated 10.5 GB 0.23 GB left (tight)
-
Wan2.1 Q4_K_MContext length 32,768 Estimated 9.16 GB 1.58 GB left
-
Context length 8,192 Estimated 10.5 GB 0.23 GB left (tight)
-
Context length 32,768 Estimated 8.70 GB 2.04 GB left
-
Context length 131,072 Estimated 8.42 GB 2.32 GB left
-
Context length 131,072 Estimated 8.07 GB 2.67 GB left
-
Context length 65,536 Estimated 8.34 GB 2.40 GB left
-
Context length 131,072 Estimated 10.7 GB 0.04 GB left (tight)
-
Hy-MT2-7B Q8_0Context length 8,192 Estimated 9.86 GB 0.88 GB left
-
Context length 8,192 Estimated 10.6 GB 0.11 GB left (tight)
-
Context length 262,144 Estimated 4.83 GB 5.90 GB left
-
Context length 8,192 Estimated 10.5 GB 0.23 GB left (tight)
-
Context length 32,768 Estimated 10.3 GB 0.46 GB left (tight)
-
Context length 65,536 Estimated 10.6 GB 0.13 GB left (tight)
-
Sharp-Spark-X2.5-4B Q6_K_XLContext length 131,072 Estimated 9.36 GB 1.37 GB left
-
AdaVaR-3B F16Context length 65,536 Estimated 9.44 GB 1.30 GB left
With 12 GB
One entry per model, up to 30, sorted by popularity. For each model we show the largest (most precise) quantization that fits in this memory with a context of at least 8K tokens. The context length is the longest one that quantization fits (for models whose maximum is under 8K, such as embedding models, that maximum). Models that are not for chat or text generation are labelled with their use. "Left" is the memory minus the estimate; under 5% of the memory is marked "tight" (not a guarantee that it fits).
-
Context length 8,192 Estimated 12.9 GB 0.00 GB left (tight)
-
Ornith-1.5-9B Q8_0Context length 32,768 Estimated 12.0 GB 0.90 GB left
-
Context length 8,192 Estimated 12.8 GB 0.12 GB left (tight)
-
Context length 32,768 Estimated 12.8 GB 0.09 GB left (tight)
-
Qwen3.8-27B-Uncensored IQ2_MContext length 8,192 Estimated 12.4 GB 0.52 GB left (tight)
-
Ornith-1.0-9B Q8_0Context length 65,536 Estimated 12.8 GB 0.08 GB left (tight)
-
Context length 512 Estimated 2.41 GB 10.5 GB left
-
Context length 16,384 Estimated 12.8 GB 0.07 GB left (tight)
-
Context length 16,384 Estimated 12.4 GB 0.51 GB left (tight)
-
Context length 16,384 Estimated 12.0 GB 0.91 GB left
-
Context length 8,192 Estimated 12.4 GB 0.47 GB left (tight)
-
Context length 8,192 Estimated 12.8 GB 0.04 GB left (tight)
-
Context length 8,192 Estimated 12.9 GB 0.00 GB left (tight)
-
Context length 32,768 Estimated 12.0 GB 0.91 GB left
-
Context length 65,536 Estimated 12.8 GB 0.11 GB left (tight)
-
Context length 8,192 Estimated 12.8 GB 0.08 GB left (tight)
-
Ternary-Bonsai-2-27B Q2_0Context length 32,768 Estimated 11.0 GB 1.91 GB left
-
Swift-1.5-Qwen3.8-27B-GSQ-RCO IQ3_XXSContext length 16,384 Estimated 12.7 GB 0.17 GB left (tight)
-
Context length 65,536 Estimated 11.9 GB 0.97 GB left
-
Context length 16,384 Estimated 12.7 GB 0.17 GB left (tight)
-
Wan2.1 IQ2_MContext length 8,192 Estimated 12.4 GB 0.49 GB left (tight)
-
Context length 16,384 Estimated 12.7 GB 0.17 GB left (tight)
-
Context length 65,536 Estimated 12.5 GB 0.43 GB left (tight)
-
Context length 4,096 Estimated 12.9 GB 0.03 GB left (tight)
-
Context length 131,072 Estimated 8.42 GB 4.47 GB left
-
Context length 262,144 Estimated 12.4 GB 0.52 GB left (tight)
-
Context length 65,536 Estimated 8.34 GB 4.55 GB left
-
Context length 262,144 Estimated 12.8 GB 0.04 GB left (tight)
-
Hy-MT2-7B Q8_0Context length 16,384 Estimated 10.9 GB 1.96 GB left
-
Context length 16,384 Estimated 12.7 GB 0.17 GB left (tight)
Frequently asked questions
- Which local LLMs run on GeForce RTX 3080?
- On this site, 753 models have a quantization that fits in 12 GB at 8,192 tokens (estimated). Popular ones include Qwen3.8-27B, Ornith-1.5-9B, Ornith-1.5-35B-A3B, Huihui-Qwen3.8-27B-abliterated, gemma-4-12B-it.
- What is the largest model that runs on GeForce RTX 3080?
- Among the models listed here, Qwen3.8-Flash-Next-FP4-Local-Strata (about 51.2B); BF16 fits in 12 GB (estimated at 8,192 tokens).