Unverified50% confidenceDecision RuleExact time
The model showed no speed advantage in TP4 mode; measured performance fell between TP1 and TP2, with TP2 considered the gold-standard configuration.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedGPT-4-level models typically achieve 30–80 TPS on consumer-grade APIs, while optimized small models or dedicated inference hardware can reach 200+ TPS.66% similarUnverifiedT3MP3ST's lateral movement and privilege escalation features are still in an experimental stage.64% similarUnverifiedFlagship models (such as GPT-4o and Claude 3.5 Sonnet) and lightweight models (such as GPT-4o-mini and Claude Haiku) can differ in price by 10–20x, yet the performance gap on a large number of routine tasks is not significant.61% similarUnverified主流商业模型如GPT-4o的输出速度通常在50-100 TPS区间60% similarUnverifiedGPT-4o achieves approximately a 2x improvement in inference speed compared to GPT-4 Turbo.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/904304API
curl https://kongchang.com/api/v1/knowledge/claims/904304MCP
get_claim(id=904304)