Akashic Computer

VRAM: 48 GB

Capacity filter

CustomGB

Only show runnable

Capacity estimate only. Runtime support remains listed per artifact.

Qwen

Alibaba / Apache-2.0

Release

6 releases. Hover or focus a release to preview its details.

Model	Runtimes	VRAM
Qwen 3.6 27BInstructBF16QwenOfficialVerifiedQwen/Qwen3.6-27B	vLLMSGLangTransformers	59–68 GB
Qwen 3.6 27BInstructFP8QwenOfficialVerifiedQwen/Qwen3.6-27B-FP8	vLLMSGLangTensorRT-LLM	30–35 GB
Qwen 3.6 27BInstructNVFP4NVIDIAVendorVerifiednvidia/Qwen3.6-27B-NVFP4	TensorRT-LLMvLLM	17–20 GB
Qwen 3.6 35BInstruct3B activeBF16QwenOfficialVerifiedQwen/Qwen3.6-35B-A3B	vLLMSGLangTransformers	77–89 GB
Qwen 3.6 35BInstruct3B activeFP8QwenOfficialVerifiedQwen/Qwen3.6-35B-A3B-FP8	vLLMSGLangTensorRT-LLM	39–45 GB
Qwen 3.6 35BInstruct3B activeNVFP4NVIDIAVendorVerifiednvidia/Qwen3.6-35B-A3B-NVFP4	TensorRT-LLMvLLM	22–25 GB
Qwen 3.5 0.8BInstructBF16QwenOfficialVerifiedQwen/Qwen3.5-0.8B	vLLMSGLangTransformers	2–3 GB
Qwen 3.5 0.8BBaseBF16QwenOfficialVerifiedQwen/Qwen3.5-0.8B-Base	vLLMSGLangTransformers	2–3 GB
Qwen 3.5 2BInstructBF16QwenOfficialVerifiedQwen/Qwen3.5-2B	vLLMSGLangTransformers	4–5 GB
Qwen 3.5 2BBaseBF16QwenOfficialVerifiedQwen/Qwen3.5-2B-Base	vLLMSGLangTransformers	4–5 GB
Qwen 3.5 4BInstructBF16QwenOfficialVerifiedQwen/Qwen3.5-4B	vLLMSGLangTransformers	9–10 GB
Qwen 3.5 4BBaseBF16QwenOfficialVerifiedQwen/Qwen3.5-4B-Base	vLLMSGLangTransformers	9–10 GB
Qwen 3.5 9BInstructBF16QwenOfficialVerifiedQwen/Qwen3.5-9B	vLLMSGLangTransformers	20–23 GB
Qwen 3.5 9BBaseBF16QwenOfficialVerifiedQwen/Qwen3.5-9B-Base	vLLMSGLangTransformers	20–23 GB
Qwen 3.5 27BInstructBF16QwenOfficialVerifiedQwen/Qwen3.5-27B	vLLMSGLangTransformers	59–68 GB
Qwen 3.5 27BInstructFP8QwenOfficialVerifiedQwen/Qwen3.5-27B-FP8	vLLMSGLangTensorRT-LLM	30–35 GB
Qwen 3.5 27BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen3.5-27B-GPTQ-Int4	vLLMSGLang	17–20 GB
Qwen 3.5 35BInstruct3B activeBF16QwenOfficialVerifiedQwen/Qwen3.5-35B-A3B	vLLMSGLangTransformers	77–89 GB
Qwen 3.5 35BInstruct3B activeFP8QwenOfficialVerifiedQwen/Qwen3.5-35B-A3B-FP8	vLLMSGLangTensorRT-LLM	39–45 GB
Qwen 3.5 35BInstruct3B activeGPTQ INT4QwenOfficialVerifiedQwen/Qwen3.5-35B-A3B-GPTQ-Int4	vLLMSGLang	22–25 GB
Qwen 3.5 35BBase3B activeBF16QwenOfficialVerifiedQwen/Qwen3.5-35B-A3B-Base	vLLMSGLangTransformers	77–89 GB
Qwen 3.5 122BInstruct10B activeBF16QwenOfficialVerifiedQwen/Qwen3.5-122B-A10B	vLLMSGLangTransformers	268–308 GB
Qwen 3.5 122BInstruct10B activeFP8QwenOfficialVerifiedQwen/Qwen3.5-122B-A10B-FP8	vLLMSGLangTensorRT-LLM	134–154 GB
Qwen 3.5 122BInstruct10B activeGPTQ INT4QwenOfficialVerifiedQwen/Qwen3.5-122B-A10B-GPTQ-Int4	vLLMSGLang	76–87 GB
Qwen 3.5 122BInstruct10B activeNVFP4NVIDIAVendorVerifiednvidia/Qwen3.5-122B-A10B-NVFP4	TensorRT-LLMvLLM	76–87 GB
Qwen 3.5 397BInstruct17B activeBF16QwenOfficialVerifiedQwen/Qwen3.5-397B-A17B	vLLMSGLangTransformers	873–1004 GB
Qwen 3.5 397BInstruct17B activeFP8QwenOfficialVerifiedQwen/Qwen3.5-397B-A17B-FP8	vLLMSGLangTensorRT-LLM	437–503 GB
Qwen 3.5 397BInstruct17B activeGPTQ INT4QwenOfficialVerifiedQwen/Qwen3.5-397B-A17B-GPTQ-Int4	vLLMSGLang	246–283 GB
Qwen 3.5 397BInstruct17B activeNVFP4NVIDIAVendorVerifiednvidia/Qwen3.5-397B-A17B-NVFP4	TensorRT-LLMvLLM	246–283 GB
Qwen 3 Coder 30BInstruct3B activeBF16QwenOfficialVerifiedQwen/Qwen3-Coder-30B-A3B-Instruct	vLLMSGLangTransformers	66–76 GB
Qwen 3 Coder 30BInstruct3B activeFP8QwenOfficialVerifiedQwen/Qwen3-Coder-30B-A3B-Instruct-FP8	vLLMSGLangTensorRT-LLM	33–38 GB
Qwen 3 Coder 480BInstruct35B activeBF16QwenOfficialVerifiedQwen/Qwen3-Coder-480B-A35B-Instruct	vLLMSGLangTransformers	1056–1214 GB
Qwen 3 Coder 480BInstruct35B activeFP8QwenOfficialVerifiedQwen/Qwen3-Coder-480B-A35B-Instruct-FP8	vLLMSGLangTensorRT-LLM	528–607 GB
Qwen 3 Coder 480BInstruct35B activeNVFP4NVIDIAVendorVerifiednvidia/Qwen3-Coder-480B-A35B-Instruct-NVFP4	TensorRT-LLMvLLM	298–343 GB
Qwen 3 Coder NextInstructBF16QwenOfficialVerifiedQwen/Qwen3-Coder-Next	vLLMSGLangTransformers	176–202 GB
Qwen 3 Coder NextInstructFP8QwenOfficialVerifiedQwen/Qwen3-Coder-Next-FP8	vLLMSGLangTensorRT-LLM	88–101 GB
Qwen 3 Coder NextInstructGGUFQwenOfficialVerifiedQwen/Qwen3-Coder-Next-GGUF	llama.cppOllamaLM Studio	53–61 GB
Qwen 3 Coder NextInstructNVFP4NVIDIAVendorVerifiednvidia/Qwen3-Next-80B-A3B-Instruct-NVFP4	TensorRT-LLMvLLM	50–57 GB
Qwen 3 Coder NextBaseBF16QwenOfficialVerifiedQwen/Qwen3-Coder-Next-Base	vLLMSGLangTransformers	176–202 GB
Qwen 3 0.6BInstructBF16QwenOfficialVerifiedQwen/Qwen3-0.6B	vLLMSGLangTransformers	2–3 GB
Qwen 3 0.6BInstructFP8QwenOfficialVerifiedQwen/Qwen3-0.6B-FP8	vLLMSGLangTransformers	2–3 GB
Qwen 3 0.6BInstructGPTQ INT8QwenOfficialVerifiedQwen/Qwen3-0.6B-GPTQ-Int8	vLLMSGLang	3–4 GB
Qwen 3 0.6BInstructGGUFQwenOfficialVerifiedQwen/Qwen3-0.6B-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 3 0.6BBaseBF16QwenOfficialVerifiedQwen/Qwen3-0.6B-Base	vLLMSGLangTransformers	2–3 GB
Qwen 3 1.7BInstructBF16QwenOfficialVerifiedQwen/Qwen3-1.7B	vLLMSGLangTransformers	4–4 GB
Qwen 3 1.7BInstructFP8QwenOfficialVerifiedQwen/Qwen3-1.7B-FP8	vLLMSGLangTransformers	2–3 GB
Qwen 3 1.7BInstructGPTQ INT8QwenOfficialVerifiedQwen/Qwen3-1.7B-GPTQ-Int8	vLLMSGLang	4–5 GB
Qwen 3 1.7BInstructGGUFQwenOfficialVerifiedQwen/Qwen3-1.7B-GGUF	llama.cppOllamaLM Studio	3–4 GB
Qwen 3 1.7BBaseBF16QwenOfficialVerifiedQwen/Qwen3-1.7B-Base	vLLMSGLangTransformers	4–4 GB
Qwen 3 4BInstructBF16QwenOfficialVerifiedQwen/Qwen3-4B	vLLMSGLangTransformers	9–10 GB
Qwen 3 4BInstructFP8QwenOfficialVerifiedQwen/Qwen3-4B-FP8	vLLMSGLangTransformers	4–5 GB
Qwen 3 4BInstructAWQ INT4QwenOfficialVerifiedQwen/Qwen3-4B-AWQ	vLLMSGLang	4–6 GB
Qwen 3 4BInstructGGUFQwenOfficialVerifiedQwen/Qwen3-4B-GGUF	llama.cppOllamaLM Studio	5–6 GB
Qwen 3 4BBaseBF16QwenOfficialVerifiedQwen/Qwen3-4B-Base	vLLMSGLangTransformers	9–10 GB
Qwen 3 8BInstructBF16QwenOfficialVerifiedQwen/Qwen3-8B	vLLMSGLangTransformers	18–20 GB
Qwen 3 8BInstructFP8QwenOfficialVerifiedQwen/Qwen3-8B-FP8	vLLMSGLangTransformers	9–10 GB
Qwen 3 8BInstructFP8NVIDIAVendorVerifiednvidia/Qwen3-8B-FP8	vLLMSGLangTransformers	9–10 GB
Qwen 3 8BInstructNVFP4NVIDIAVendorVerifiednvidia/Qwen3-8B-NVFP4	TensorRT-LLMvLLM	5–6 GB
Qwen 3 8BInstructAWQ INT4QwenOfficialVerifiedQwen/Qwen3-8B-AWQ	vLLMSGLang	7–9 GB
Qwen 3 8BInstructGGUFQwenOfficialVerifiedQwen/Qwen3-8B-GGUF	llama.cppOllamaLM Studio	7–9 GB
Qwen 3 8BBaseBF16QwenOfficialVerifiedQwen/Qwen3-8B-Base	vLLMSGLangTransformers	18–20 GB
Qwen 3 14BInstructBF16QwenOfficialVerifiedQwen/Qwen3-14B	vLLMSGLangTransformers	31–35 GB
Qwen 3 14BInstructFP8QwenOfficialVerifiedQwen/Qwen3-14B-FP8	vLLMSGLangTransformers	15–18 GB
Qwen 3 14BInstructFP8NVIDIAVendorVerifiednvidia/Qwen3-14B-FP8	vLLMSGLangTransformers	15–18 GB
Qwen 3 14BInstructNVFP4NVIDIAVendorVerifiednvidia/Qwen3-14B-NVFP4	TensorRT-LLMvLLM	9–10 GB
Qwen 3 14BInstructAWQ INT4QwenOfficialVerifiedQwen/Qwen3-14B-AWQ	vLLMSGLang	11–13 GB
Qwen 3 14BInstructGGUFQwenOfficialVerifiedQwen/Qwen3-14B-GGUF	llama.cppOllamaLM Studio	11–14 GB
Qwen 3 14BBaseBF16QwenOfficialVerifiedQwen/Qwen3-14B-Base	vLLMSGLangTransformers	31–35 GB
Qwen 3 30BInstruct3B activeBF16QwenOfficialVerifiedQwen/Qwen3-30B-A3B	vLLMSGLangTransformers	66–76 GB
Qwen 3 30BInstruct3B activeFP8QwenOfficialVerifiedQwen/Qwen3-30B-A3B-FP8	vLLMSGLangTransformers	33–38 GB
Qwen 3 30BInstruct3B activeNVFP4NVIDIAVendorVerifiednvidia/Qwen3-30B-A3B-NVFP4	TensorRT-LLMvLLM	19–21 GB
Qwen 3 30BInstruct3B activeGPTQ INT4QwenOfficialVerifiedQwen/Qwen3-30B-A3B-GPTQ-Int4	vLLMSGLang	21–25 GB
Qwen 3 30BInstruct3B activeGGUFQwenOfficialVerifiedQwen/Qwen3-30B-A3B-GGUF	llama.cppOllamaLM Studio	22–27 GB
Qwen 3 30BBase3B activeBF16QwenOfficialVerifiedQwen/Qwen3-30B-A3B-Base	vLLMSGLangTransformers	66–76 GB
Qwen 3 32BInstructBF16QwenOfficialVerifiedQwen/Qwen3-32B	vLLMSGLangTransformers	70–81 GB
Qwen 3 32BInstructFP8QwenOfficialVerifiedQwen/Qwen3-32B-FP8	vLLMSGLangTransformers	35–40 GB
Qwen 3 32BInstructNVFP4NVIDIAVendorVerifiednvidia/Qwen3-32B-NVFP4	TensorRT-LLMvLLM	20–23 GB
Qwen 3 32BInstructAWQ INT4QwenOfficialVerifiedQwen/Qwen3-32B-AWQ	vLLMSGLang	22–27 GB
Qwen 3 32BInstructGGUFQwenOfficialVerifiedQwen/Qwen3-32B-GGUF	llama.cppOllamaLM Studio	23–29 GB
Qwen 3 235BInstruct22B activeBF16QwenOfficialVerifiedQwen/Qwen3-235B-A22B	vLLMSGLangTransformers	517–595 GB
Qwen 3 235BInstruct22B activeFP8QwenOfficialVerifiedQwen/Qwen3-235B-A22B-FP8	vLLMSGLangTransformers	259–297 GB
Qwen 3 235BInstruct22B activeFP8NVIDIAVendorVerifiednvidia/Qwen3-235B-A22B-FP8	vLLMSGLangTransformers	259–297 GB
Qwen 3 235BInstruct22B activeNVFP4NVIDIAVendorVerifiednvidia/Qwen3-235B-A22B-NVFP4	TensorRT-LLMvLLM	146–168 GB
Qwen 3 235BInstruct22B activeGPTQ INT4QwenOfficialVerifiedQwen/Qwen3-235B-A22B-GPTQ-Int4	vLLMSGLang	148–177 GB
Qwen 3 235BInstruct22B activeGGUFQwenOfficialVerifiedQwen/Qwen3-235B-A22B-GGUF	llama.cppOllamaLM Studio	157–191 GB
Qwen 2.5 Coder 0.5BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-0.5B-Instruct	vLLMSGLangTransformers	2–3 GB
Qwen 2.5 Coder 0.5BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-Coder-0.5B-Instruct-AWQ	vLLMSGLang	2–3 GB
Qwen 2.5 Coder 0.5BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-Coder-0.5B-Instruct-GPTQ-Int4	vLLMSGLang	2–3 GB
Qwen 2.5 Coder 0.5BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-Coder-0.5B-Instruct-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 2.5 Coder 0.5BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-0.5B	vLLMSGLangTransformers	2–3 GB
Qwen 2.5 Coder 1.5BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-1.5B-Instruct	vLLMSGLangTransformers	3–3 GB
Qwen 2.5 Coder 1.5BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-Coder-1.5B-Instruct-AWQ	vLLMSGLang	2–3 GB
Qwen 2.5 Coder 1.5BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-Coder-1.5B-Instruct-GPTQ-Int4	vLLMSGLang	2–3 GB
Qwen 2.5 Coder 1.5BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-Coder-1.5B-Instruct-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 2.5 Coder 1.5BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-1.5B	vLLMSGLangTransformers	3–3 GB
Qwen 2.5 Coder 3BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-3B-Instruct	vLLMSGLangTransformers	7–8 GB
Qwen 2.5 Coder 3BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-Coder-3B-Instruct-AWQ	vLLMSGLang	2–3 GB
Qwen 2.5 Coder 3BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-Coder-3B-Instruct-GPTQ-Int4	vLLMSGLang	2–3 GB
Qwen 2.5 Coder 3BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-Coder-3B-Instruct-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 2.5 Coder 3BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-3B	vLLMSGLangTransformers	7–8 GB
Qwen 2.5 Coder 7BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-7B-Instruct	vLLMSGLangTransformers	15–17 GB
Qwen 2.5 Coder 7BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-Coder-7B-Instruct-AWQ	vLLMSGLang	4–5 GB
Qwen 2.5 Coder 7BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-Coder-7B-Instruct-GPTQ-Int4	vLLMSGLang	4–5 GB
Qwen 2.5 Coder 7BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-Coder-7B-Instruct-GGUF	llama.cppOllamaLM Studio	5–6 GB
Qwen 2.5 Coder 7BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-7B	vLLMSGLangTransformers	15–17 GB
Qwen 2.5 Coder 14BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-14B-Instruct	vLLMSGLangTransformers	31–36 GB
Qwen 2.5 Coder 14BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-Coder-14B-Instruct-AWQ	vLLMSGLang	9–10 GB
Qwen 2.5 Coder 14BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-Coder-14B-Instruct-GPTQ-Int4	vLLMSGLang	9–10 GB
Qwen 2.5 Coder 14BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-Coder-14B-Instruct-GGUF	llama.cppOllamaLM Studio	9–10 GB
Qwen 2.5 Coder 14BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-14B	vLLMSGLangTransformers	31–36 GB
Qwen 2.5 Coder 32BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-32B-Instruct	vLLMSGLangTransformers	70–81 GB
Qwen 2.5 Coder 32BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-Coder-32B-Instruct-AWQ	vLLMSGLang	20–23 GB
Qwen 2.5 Coder 32BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-Coder-32B-Instruct-GPTQ-Int4	vLLMSGLang	20–23 GB
Qwen 2.5 Coder 32BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-Coder-32B-Instruct-GGUF	llama.cppOllamaLM Studio	21–24 GB
Qwen 2.5 Coder 32BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-Coder-32B	vLLMSGLangTransformers	70–81 GB
Qwen 2.5 0.5BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-0.5B-Instruct	vLLMSGLangTransformers	2–3 GB
Qwen 2.5 0.5BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-0.5B-Instruct-AWQ	vLLMSGLang	2–3 GB
Qwen 2.5 0.5BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-0.5B-Instruct-GPTQ-Int4	vLLMSGLang	2–3 GB
Qwen 2.5 0.5BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-0.5B-Instruct-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 2.5 0.5BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-0.5B	vLLMSGLangTransformers	2–3 GB
Qwen 2.5 1.5BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-1.5B-Instruct	vLLMSGLangTransformers	3–3 GB
Qwen 2.5 1.5BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-1.5B-Instruct-AWQ	vLLMSGLang	2–3 GB
Qwen 2.5 1.5BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-1.5B-Instruct-GPTQ-Int4	vLLMSGLang	2–3 GB
Qwen 2.5 1.5BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-1.5B-Instruct-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 2.5 1.5BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-1.5B	vLLMSGLangTransformers	3–3 GB
Qwen 2.5 3BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-3B-Instruct	vLLMSGLangTransformers	7–8 GB
Qwen 2.5 3BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-3B-Instruct-AWQ	vLLMSGLang	2–3 GB
Qwen 2.5 3BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-3B-Instruct-GPTQ-Int4	vLLMSGLang	2–3 GB
Qwen 2.5 3BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-3B-Instruct-GGUF	llama.cppOllamaLM Studio	2–3 GB
Qwen 2.5 3BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-3B	vLLMSGLangTransformers	7–8 GB
Qwen 2.5 7BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-7B-Instruct	vLLMSGLangTransformers	15–17 GB
Qwen 2.5 7BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-7B-Instruct-AWQ	vLLMSGLang	4–5 GB
Qwen 2.5 7BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-7B-Instruct-GPTQ-Int4	vLLMSGLang	4–5 GB
Qwen 2.5 7BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-7B-Instruct-GGUF	llama.cppOllamaLM Studio	5–6 GB
Qwen 2.5 7BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-7B	vLLMSGLangTransformers	15–17 GB
Qwen 2.5 14BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-14B-Instruct	vLLMSGLangTransformers	31–36 GB
Qwen 2.5 14BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-14B-Instruct-AWQ	vLLMSGLang	9–10 GB
Qwen 2.5 14BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-14B-Instruct-GPTQ-Int4	vLLMSGLang	9–10 GB
Qwen 2.5 14BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-14B-Instruct-GGUF	llama.cppOllamaLM Studio	9–10 GB
Qwen 2.5 14BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-14B	vLLMSGLangTransformers	31–36 GB
Qwen 2.5 32BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-32B-Instruct	vLLMSGLangTransformers	70–81 GB
Qwen 2.5 32BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-32B-Instruct-AWQ	vLLMSGLang	20–23 GB
Qwen 2.5 32BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-32B-Instruct-GPTQ-Int4	vLLMSGLang	20–23 GB
Qwen 2.5 32BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-32B-Instruct-GGUF	llama.cppOllamaLM Studio	21–24 GB
Qwen 2.5 32BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-32B	vLLMSGLangTransformers	70–81 GB
Qwen 2.5 72BInstructBF16QwenOfficialVerifiedQwen/Qwen2.5-72B-Instruct	vLLMSGLangTransformers	158–182 GB
Qwen 2.5 72BInstructAWQQwenOfficialVerifiedQwen/Qwen2.5-72B-Instruct-AWQ	vLLMSGLang	45–52 GB
Qwen 2.5 72BInstructGPTQ INT4QwenOfficialVerifiedQwen/Qwen2.5-72B-Instruct-GPTQ-Int4	vLLMSGLang	45–52 GB
Qwen 2.5 72BInstructGGUFQwenOfficialVerifiedQwen/Qwen2.5-72B-Instruct-GGUF	llama.cppOllamaLM Studio	48–55 GB
Qwen 2.5 72BBaseBF16QwenOfficialVerifiedQwen/Qwen2.5-72B	vLLMSGLangTransformers	158–182 GB