Expired50% confidenceOpinionExact time
For 8 GB VRAM, Llama 3.1 8B Instruct, Qwen2.5 7B Instruct, and Mistral 7B with Q4_K_M quantization are recommended.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/16/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/14/2026(expired)
Sources
Related Claims
Unverified8GB of VRAM is sufficient to run Q4/Q5 quantized versions of 7B models (all layers on GPU), while 12GB of VRAM can run Q4 quantized versions of 14B models or Q8 versions of 7B models.79% similarVerified8 GB of VRAM is recommended for running 7B models (Q4 quantization), 16 GB for 7B to 14B models, 24 GB for 14B to 32B models, and 32 GB for 32B to 70B models (Q4 quantization)74% similarUnverifiedThe developer provides five quantization formats: INT8 ConvRot (4.5 GB), FP8 E4M3 (4.2 GB), bf16 (8.3 GB), NVFP4 (2.9 GB), and MXFP8 (4.7 GB).74% similarUnverifiedModels in the 3B–4B range, such as Qwen2.5 3B or Llama 3.2 3B, require only 2 to 2.5GB of VRAM under Q4 quantization.73% similarUnverifiedQwen2 7B模型（70亿参数）运行约需6GB显存73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/536144API
curl https://kongchang.com/api/v1/knowledge/claims/536144MCP
get_claim(id=536144)