Unverified50% confidenceTradeoffExact time
IQ4_XS与Q4_K_M速度几乎一致，但IQ4_XS体积更小可节省约2GB空间用于扩展上下文，因此性价比更高
1
Sources
50%
Confidence
Long-term
Relevance
8/20/2026
First Seen
Sources
Related Claims
VerifiedQ4_K_M uses mixed precision, storing some critical weights at 6-bit and the remainder at 4-bit, with an overall average of approximately 4.5-bit.72% similarUnverified一个32K上下文窗口在Q4量化模型上可能额外占用2-4GB内存69% similarUnverifiedQ4_K_M is the de facto standard choice for users with 8 GB of VRAM, significantly improving output quality compared to Q4_0 with almost no increase in VRAM usage.69% similarUnverifiedAfter Gemma 4 QAT optimization, the minimum memory footprint can be reduced to 1GB68% similarVerified4-bit quantization (such as Q4_K_M) can typically compress the memory footprint of a 3B model from approximately 6GB to approximately 2GB, with a precision loss of less than 2–3%.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/778777API
curl https://kongchang.com/api/v1/knowledge/claims/778777MCP
get_claim(id=778777)