cm-llm-manager apollo

Qwen3.8-27B-UD-IQ4_XS.gguf

llama-rocm-qwen38-27b-mtp · alias Qwen3.8-27B · 262,144 ctx · ready

Stacks

Stack	Model	Context	State
`llama-rocm-gemma4`	gemma-4-26B-A4B-it-UD-Q4_K_M.gguf gemma4-26b-a4b	256,000	ready
`llama-rocm-gpt-oss`	gpt-oss-20b-UD-Q4_K_XL.gguf gpt-oss	196,000	ready
`llama-rocm-qwen36-35b` no_alias, presence_penalty_underscore	Qwen3.6-35B-A3B-UD-IQ4_XS.gguf	128,000	ready
`llama-rocm-qwen36-35b-mtp` no_alias	Qwen3.6-35B-A3B-MTP-UD-IQ4_XS.gguf	262,144	ready
`llama-rocm-qwen38-27b` no_alias	Qwen3.8-27B-UD-Q4_K_XL.gguf	65,536	ready
`llama-rocm-qwen38-27b-mtp`	Qwen3.8-27B-UD-IQ4_XS.gguf Qwen3.8-27B	278,528	active
`llama-rocm` models_mount_not_opt_models	gemma-4-26B-A4B-it-UD-Q4_K_M.gguf gemma4-26b-a4b	196,000	unusable
`llama-rocm-qwen36-27b` no_alias, presence_penalty_underscore models_mount_not_opt_models	Qwen3.6-27B-Q4_K_S.gguf	96,000	unusable

Sign in

Switching stops the public model for one to three minutes. Scripts send Authorization: Bearer <token> instead.

Recent switches

When	Target	Result	Took
2026-10-10T02:38:36Z	`llama-rocm-qwen38-27b-mtp`	ready	12.9s
2026-10-10T02:37:04Z	`llama-rocm-qwen38-27b-mtp`	ready	25.9s
2026-10-10T02:36:09Z	`llama-rocm-qwen38-27b`	ready	34.3s