Unverified50% confidenceFactExact time
CrofAI claimed to run Kimi K3 on RTX Pro 6000 units rented from Vast, but even with Q2_K quantization the model requires approximately 802 GiB of VRAM, while the largest 8-GPU RTX PRO 6000 machine available on Vast has only 765 GiB of VRAM.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/15/2026
First Seen
Valid until: 12/14/2026
Sources
Related Entities
Related Claims
UnverifiedAn RTX 3060 with 6 GB of VRAM can run the Flux-based Krea2 Turbo large model smoothly.72% similarUnverifiedIn actual testing, a 1-billion-parameter draft model was mounted onto Qwen3's 3.4B target model for testing on an RTX A6000 (48GB VRAM) machine.70% similarUnverifiedWhen batching 8 requests together, a single RTX PRO 6000 can achieve an aggregate throughput of approximately 220 output tok/s, whereas the old configuration of two RTX 3090s achieved only approximately 19 tok/s.70% similarUnverifiedThe UP host's test environment consisted of an RTX 5070 GPU (12GB VRAM), the Qwen3 series 9B parameter model, and a 64K context length.69% similarUnverifiedWhen fine-tuning a Llama 7B model on an RTX 4090 using QLoRA, VRAM usage is approximately 5–8GB.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/922187API
curl https://kongchang.com/api/v1/knowledge/claims/922187MCP
get_claim(id=922187)