Unverified50% confidenceBenchmarkExact time
On segments unaffected by digits and writing conventions, Sada-ASR's WER lead over Microsoft MAI-Transcribe-2 was 0.74 percentage points.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/17/2026
First Seen
Valid until: 2/15/2027
Sources
Related Entities
Related Claims
UnverifiedGPT-6 Astra peaked at 98% on Frontier Math Tier 4, and still reached 83% without enabling reasoning or using chain-of-thought.65% similarUnverifiedFable 5.2 reaches a 41% automation rate on the Remote Labor Index (Remark benchmark), whereas a year earlier Opus 4.5 was only about 3.8%, and GPT-6 Astra reached about 21% in September.62% similarUnverifiedOn the Bench 2.0 benchmark, an internal model scored 0.875 while Astra scored 0.594; the internal model's lowest reasoning tier scored 0.758, already surpassing Astra's result when running at full effort.61% similarUnverifiedCompared to GPT-6 Astra, SWE-2's score trails by a few percentage points, but its cost is only one-quarter of the latter's.61% similarUnverifiedIn DeepSeek's official comparison, when running with the same checkpoint at best effort, its minimal framework scored 72.6 while the open-source code scored 65.5.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1178047API
curl https://kongchang.com/api/v1/knowledge/claims/1178047MCP
get_claim(id=1178047)