Expired50% confidenceBenchmarkExact time
Data from Linux.do discussion threads shows that Astra consumes approximately 4–5 times the quota of Sol, and the cache hit rates of both models are as high as 97–98%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/7/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/21/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedCombining the 2.5x pricing differential and the 1.8x additional multiplier, Astra's final effective quota consumption difference relative to Sol reaches approximately 4.5 times.76% similarUnverifiedOn the Artificial Analysis coding agent index, Sol set to the highest reasoning tier scored 80, surpassing Fable 5 by 2.8 points, while using less than half the output tokens, taking less than half the time, and costing roughly one-third less.65% similarUnverifiedIn the OS World computer operation test, Sol surpassed Claude's Opus series while consuming 85% less compute.63% similarUnverifiedOn SimpleBench, Sol Pro scored 71.7% and Sol approximately 65%, still 10 percentage points below Fable.61% similarUnverifiedOn the ARC-AGI-3 benchmark, GPT-5.6 Sol scored 8%, while most other models scored below 2%; Anthropic did not run Fable due to prohibitively high costs.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/870389API
curl https://kongchang.com/api/v1/knowledge/claims/870389MCP
get_claim(id=870389)