Unverified50% confidenceBenchmarkExact time
In the short question-answering test (a 400-character input plus a small question), the version with MTP reduced first-token latency from 5.9 seconds to 4.3 seconds, increased output speed from 57.98 characters/second to 66.14 characters/second (about 14% faster), and shortened the total time to finish the answer from 85.9 seconds to 62.3 seconds (27% faster).
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/28/2026
First Seen
Valid until: 11/11/2026
Sources
Related Entities
Related Claims
UnverifiedAfter enabling MTP2, total generation time drops from 37.1 seconds to 24.7 seconds, with sustained generation speed increasing by approximately 60%.77% similarVerifiedIn testing, the p99 latency per 160-millisecond tick was 147.4–147.5 milliseconds, with zero missed deadlines across three rounds totaling 84,000 session ticks measured.75% similarUnverifiedThe full jitter strategy reduced p95 latency from 16.1 seconds to 10.5 seconds, but the last call completed at 17.1 seconds, and it sent 408 requests, 9% more than the fixed-delay strategy.75% similarUnverifiedRunning a 27B model on M1 Max with a 32,000-token prompt requires waiting more than 500 seconds before the first character is output, whereas the M5 Max takes less than one minute (approximately 40 seconds with optimized software).74% similarUnverifiedTesting the same scene and the same sentence 10 times each, DeepSeek's average response time was 3.68 seconds, M3's was 10.70 seconds, and pre-optimization M3.1's was as high as 20.35 seconds.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1106416API
curl https://kongchang.com/api/v1/knowledge/claims/1106416MCP
get_claim(id=1106416)