Unverified50% confidenceBenchmarkExact time
On M4 Max, Omni 3B has a median dialogue generation time of 171 ms and a peak process RSS of 2.57 GiB; Qwen3.5 0.8B has a median dialogue generation time of 96 ms and a peak process RSS of 1.08 GiB; Qwen3-TTS 1.7B has a synthesis time of 1.78 seconds and a peak process RSS of 3.85 GiB.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/14/2026
First Seen
Valid until: 10/28/2026
Sources
Related Entities
Related Claims
UnverifiedThe customer Quantium case shows that a coding task originally required 38 prompt interactions and took 4 days, but after using Opus 5.5 it was shortened to 11 prompts and completed within 3 hours.60% similarUnverifiedIn the author's own tests, Haiku 5.5 completed the same task in approximately 20 seconds, while Grok 4.7 took approximately 35 minutes, at roughly one-fifth the cost.60% similarUnverifiedCompleting the 30-second version involved 14 sub-agents working in coordination, with the entire process involving 779 tool calls (161 in the main conversation and 618 in sub-agents), taking approximately 2.4 hours in total.58% similarUnverified4.1 Flash completed a long-horizon programming task in 21 minutes consuming 32.1 million tokens, but exhibited three noticeable front-end bugs.58% similarUnverifiedBased on client-side observations from two demonstrations, the author estimated an average of approximately 242 tokens per second for coding/repair tasks, approximately 269 for web tasks, and a weighted average of approximately 262 tokens per second (including reasoning tokens); factoring in request wait time, the overall rate was approximately 136 tokens per second.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023263API
curl https://kongchang.com/api/v1/knowledge/claims/1023263MCP
get_claim(id=1023263)