Unverified50% confidenceEventExact time
The author developed GFX1151 Engine primarily with assistance from large models such as Kimi and ChatGPT; after independently reaching approximately 600 tokens/s, further improvement proved difficult, and the speed was ultimately raised to 800–1,000 tokens/s by reverse-engineering and referencing the Hologen Flash Server approach.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/5/2026
First Seen
Valid until: 1/3/2027
Sources
Related Entities
Related Claims
UnverifiedKling 4.0's prompt length has been expanded to 8,000 tokens.60% similarUnverifiedKolibri uses a Mixture of Experts (MoE) architecture with approximately 78 billion total parameters, activating only approximately 3.46 billion parameters per token, and is open under the Apache 2.0 license.57% similarUnverifiedOver the past week, V4.1 Flash topped the global model invocation volume rankings for the first time, with a single-week Token call volume of 15.8 trillion, a week-over-week surge of 219%.55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/978822API
curl https://kongchang.com/api/v1/knowledge/claims/978822MCP
get_claim(id=978822)