Unverified50% confidenceBenchmarkExact time
The project used two RTX 4090 GPUs each with 48 GB of VRAM to run 8 long-context requests each approaching 256K.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe RTX 5090 has 32 GB of VRAM.77% similarUnverifiedThe RTX 4090 has 24 GB of VRAM, and after loading the 27B Q4 model, there is typically enough space remaining to support a context of 30,000–40,000 tokens.75% similarUnverifiedOn an RTX 4080, the system can generate video at speeds exceeding 200 FPS, occupying only approximately 17.2 MB of VRAM during inference.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/923182API
curl https://kongchang.com/api/v1/knowledge/claims/923182MCP
get_claim(id=923182)