Unverified50% confidenceBenchmarkExact time
The Prism ML white paper reports that Bonsai 2 retains approximately 76% of the original model's score on TerminalBench 2.1 and approximately 75% on SWE-bench Verified, two long-chain benchmarks that were not included in the 98% average.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/4/2026
First Seen
Valid until: 1/2/2027
Sources
Related Entities
Related Claims
UnverifiedThe official claim is that Bonsai 2 can retain over 98% of FP16 performance on the Top-1 metric.63% similarUnverifiedAccording to the model card, Ternary Bonsai 2 is reduced to one-ninth the size of FP16 while retaining 98.2% of its intelligence performance.63% similarUnverifiedIn a needle-in-a-haystack test with a 256K context, both models performed identically, each achieving a score of 93%, with each failing only once at 100% full context depth.57% similarUnverifiedUnder the DeepSuite V1.1 benchmark, the same model's scores fluctuate from 69.8% to 74.2% across different harnesses.56% similarUnverified测试数据表明RNN头与马尔可夫头的草稿通过率差距仅在1-2%以内，而部署成本差距可达数倍55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/973453API
curl https://kongchang.com/api/v1/knowledge/claims/973453MCP
get_claim(id=973453)