# alias	hf_spec	placement	note	devices
gpt-oss-120b	ggml-org/gpt-oss-120b-GGUF	gpu0	MXFP4 native, ~60 GB, single card	0
qwen3-coder-next-80b	unsloth/Qwen3-Coder-Next-GGUF:Q8_0	gpu0	80B-A3B, Q8 ~85 GB, single card	0
qwen2.5-coder-32b	unsloth/Qwen2.5-Coder-32B-Instruct-GGUF:Q8_0	gpu0	32B dense, Q8 ~35 GB	0
devstral-small-2-24b	unsloth/Devstral-Small-2507-GGUF:Q8_0	gpu0	24B dense, Q8 ~26 GB (verify repo)	0
qwen3-235b-a22b	unsloth/Qwen3-235B-A22B-Instruct-2507-GGUF:Q4_K_M	dual	235B-A22B, Q4 ~133 GB, dual	0,1
minimax-m2.7	unsloth/MiniMax-M2.7-GGUF:UD-Q4_K_XL	dual	~230B-A10B, Q4, dual (verify tag)	0,1
glm-4.6-355b	unsloth/GLM-4.6-GGUF:UD-Q2_K_XL	dual	355B-A32B, Q2_K_XL ~135 GB, dual	0,1
qwen3-coder-480b	unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF:UD-Q2_K_XL	dual	480B-A35B, Q2 ~180 GB, dual (tight)	0,1
qwen3.6-27b	unsloth/Qwen3.6-27B-GGUF:Q8_0	gpu0	27B dense, Q8 ~30 GB, single card	0
qwen3.6-35b-a3b	unsloth/Qwen3.6-35B-A3B-GGUF:Q8_0	gpu0	35B-A3B MoE, Q8 ~38 GB, single card	0
hermes-4.3-36b	NousResearch/Hermes-4.3-36B-GGUF:Q8_0	gpu0	36B dense (Seed-OSS-36B), Q8 ~38 GB, single card	0
hermes-4-70b	unsloth/Hermes-4-70B-GGUF:Q8_0	gpu0	70B dense (Llama 3.1), Q8 ~75 GB, single card	0