Unverified50% confidenceBenchmarkExact time
In the SWE BenchPro test, Opus with medium solved 92.8% of tasks and Fable with high solved 92.3% of tasks, but Fable's cost per solved task was about 5 times that of the former.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/8/2026
First Seen
Valid until: 11/22/2026
Sources
The Ultimate Guide to Claude Models and Effort: How to Save Tokens
youtubeWestPoint 11/7/2026
Related Entities
Related Claims
UnverifiedFor Claude Opus 5.5, which can already solve 96% of tasks on its own, running it in BLM mode reduced model calls and per-task cost by about 98% and additionally solved 2 tasks.69% similarUnverifiedThe idealized optimal task-model matching achieves a solution rate of 97.6%, at a cost of only $1.88 per task.69% similarUnverifiedOn long-horizon difficult tasks, Opus rose from an 87% success rate at the low level to 97% at the x-high level, but the cost per task increased nearly 7-fold.68% similarUnverifiedIn a benchmark on 50 synthetic chore scenarios, the fine-tuned SlothDistill, compared with the base Llama-3-8B, reduced p95 decision latency from 1,420 ms to 185 ms (a 7.6x improvement), cut average output tokens from 412 to 48 (an 88.3% saving), raised direct-action accuracy from 62.4% to 97.8%, and lowered API cost per thousand tasks from $1.85 to $0.22 (88.1% cheaper).67% similarUnverifiedThe strongest configuration solved 78.2% of tasks under the strict task-level criteria and achieved a pass rate of 94.8% under the standard-level evaluation.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1147914API
curl https://kongchang.com/api/v1/knowledge/claims/1147914MCP
get_claim(id=1147914)