Unverified50% confidenceBenchmarkExact time
Published latency: RTX 4090 (bf16) 8 ms for a single question, 17 ms for a 384px image, and 475 states/s batched throughput; Apple M5 Pro (MPS) 30 ms / 62 ms / 78 states/s.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedAn RTX 4080S (Q4 quantization) used as a comparison can reach over 1,000 t/s during the preview phase.67% similarUnverifiedOn the Intel Skylake microarchitecture, the latency of 64-bit IDIV is 35–88 cycles.66% similarUnverifiedMiniMax H3 REF completes native 4096×1536 resolution image generation in approximately 1 minute on a single RTX 5060 Ti 16GB GPU with 18 sampling steps.62% similarUnverifiedThe M5 Ultra Mac Studio running Cyberpunk 2077 at ultra settings with AI super-sampling disabled achieves 178 fps at 1080P, 131 fps at 2K, and maintains 68.8 fps at 4K.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1005999API
curl https://kongchang.com/api/v1/knowledge/claims/1005999MCP
get_claim(id=1005999)