Unverified50% confidenceFactExact time
StepFun's Step 3.5 can reach an inference speed of six to seven hundred tokens per second, while by Step 5 the speed drops to under one hundred per second, yet capability has not improved noticeably.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/5/2026
First Seen
Valid until: 2/3/2027
Sources
Related Entities
Related Claims
UnverifiedIf the optimizer itself is 1.5 times faster, in the case where the optimizer accounts for 10%, the end-to-end speedup is about 1/(0.9+0.1/1.5)≈1.03, that is, only about 3%.69% similarUnverifiedAfter restoring flushing, the Claude candidate achieved a completion speedup of 2.75× and a time-to-first-byte speedup of 2.80×, while Gemini achieved a completion speedup of 3.44× and a time-to-first-byte speedup of 3.54×, and GPT-5.4 achieved a completion speedup of 2.16× and a time-to-first-byte speedup of 2.19×.69% similarUnverifiedThe three-level cache improved generation speed under Q4 weights from 1.83 tokens per second to 8.89, an improvement of nearly 5x.65% similarUnverifiedAcross nine revised runs, the median completion speedup was 5.35× for Gemini, 3.69× for Claude, and 1.00× for GPT-5.4, with all three passing automated checks.65% similarUnverifiedAccording to Table 3 of the paper, Gemma-4-12B-it achieves a single-step accuracy of 74.4% and an end-to-end task success rate of only 5.0%.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1140878API
curl https://kongchang.com/api/v1/knowledge/claims/1140878MCP
get_claim(id=1140878)