Unverified50% confidenceFactExact time
The complete training process for Supra2-IMG was completed on a single H100 on the Runpod platform, taking less than 10 hours.
1
Sources
50%
Confidence
Long-term
Relevance
9/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe project's policy was trained from scratch using mjlab (a MuJoCo wrapper tool) and converged in approximately 4 hours on a single RTX 3080.65% similarUnverifiedUsing Omnigent's cross-model workflow, the actual time from scratch to a working run took less than 10 minutes.64% similarUnverified在OSWorld 2.0最长的工作流中AI完成率几乎降至0，完成率对任务长度极其敏感63% similarUnverifiedThe solution achieves a stable rate of approximately 40 tok/s on prose/text generation tasks and approximately 75 tok/s on programming tasks, with speed sustained throughout the entire 1M context without significant degradation.60% similarUnverifiedUsing AI agents to generate JMeter scripts reduced the time required from approximately 1 hour to approximately 2 minutes.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/939173API
curl https://kongchang.com/api/v1/knowledge/claims/939173MCP
get_claim(id=939173)