Unverified50% confidenceBenchmarkExact time
In the Automation Bench test, the score improved from Cell's 8.1 to Astra's 41.4, and the CAD task score improved from 83.3 to 95.9.
1
Sources
50%
Confidence
Long-term
Relevance
9/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAcross all four groups and eight rounds of testing, Astra showed significant improvement over 5.6 in front-end design, game replication, and 3D interaction tasks, while 5.6's advantage lay in its speed.64% similarUnverifiedIn the Terminal Bench 4.0 highest-accuracy test, Fable 5.1 achieved an accuracy of 55.8% at a cost of $19.50, while Astra achieved an accuracy of 56.7% at a cost of $10.35.62% similarUnverified在 Terminal Bench 2.1 测试中，AutoPrompt Scale 使 OpenCode 在 89 个任务中的失败数从 29 降到 16，多解决 13 题，分数提升 14.61 个百分点62% similarUnverifiedIn the LongMemEval benchmark, BM25 accuracy improved from 78.8% to 85.2% with 5 chunks.60% similarUnverifiedK2.7 Code improved by 31.5% over K2.6 on the MLS Bench Lite benchmark (from 26.7 to 35.1), the largest improvement across all tests.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/938899API
curl https://kongchang.com/api/v1/knowledge/claims/938899MCP
get_claim(id=938899)