Unverified50% confidenceFactExact time
Bonsai 2 uses the PQ2_0 ternary quantization format, with a model size of 7.2GB.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/18/2026
First Seen
Valid until: 12/17/2026
Sources
Related Entities
Related Claims
UnverifiedThe Bonsai ternary version is approximately 6GB at 1.71 bits and retains approximately 97% of the base model's quality; the 1-bit version is 3.9GB and retains approximately 90% of quality73% similarUnverifiedA size of 7.2GB means Bonsai 2 can run on a single RTX 3060 at a reasonable tps.72% similarUnverifiedBonsai is based on Qwen 3's full-size 27B model70% similarUnverifiedBonsai 2, released by PrismML, is a heavily quantized version based on Qwen 3.8 27B.68% similarUnverifiedBonsai compresses a 27B model from 54GB at full precision down to 3.9GB66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/933012API
curl https://kongchang.com/api/v1/knowledge/claims/933012MCP
get_claim(id=933012)