Exact artifacts
Validated Ollama variants
Every entry is pinned to a quantization, blob size and digest.
Q4_K_M · 2.32 GB
Phi-4 mini
phi4-mini:3.8b-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 2.33 GBQwen3
qwen3:4b-instruct-2507-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 4.36 GBQwen2.5 Coder
qwen2.5-coder:7b-instruct-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 6.96 GBMistral Nemo
mistral-nemo:12b-instruct-2407-q4_K_M
Compare NVIDIA GPUs → MXFP4 · 12.8 GBGPT-OSS
gpt-oss:20b
Compare NVIDIA GPUs → Q4_K_M · 3.16 GBQwen3.5 4B
qwen3.5:4b-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 6.14 GBQwen3.5 9B
qwen3.5:9b-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 16.2 GBQwen3.5 27B
qwen3.5:27b-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 22.2 GBQwen3.5 35B-A3B
qwen3.5:35b-a3b-q4_K_M
Compare NVIDIA GPUs → Q4_K_M · 4.87 GBDeepSeek-R1 8B
deepseek-r1:8b
Compare NVIDIA GPUs → Q4_K_M · 8.37 GBDeepSeek-R1 14B
deepseek-r1:14b
Compare NVIDIA GPUs → Q4_K_M · 18.5 GBDeepSeek-R1 32B
deepseek-r1:32b
Compare NVIDIA GPUs → Q4_K_M · 3.11 GBGemma 3 4B
gemma3:4b
Compare NVIDIA GPUs → Q4_K_M · 7.59 GBGemma 3 12B
gemma3:12b
Compare NVIDIA GPUs → Q4_K_M · 16.2 GBGemma 3 27B
gemma3:27b
Compare NVIDIA GPUs →