Model Anatomy See what's inside open-weight AI models

Can I run Qwen3-235B-A22B-Thinking-2507-FP8 on an NVIDIA B200?

Probably, at 4-bit

The published weights are too big, but a 4-bit version (about 132 GB) would fit, with room for ≈ 151K tokens of context. The site hasn't analyzed a GGUF of Qwen3-235B-A22B-Thinking-2507-FP8 yet.

Full analysis of Qwen3-235B-A22B-Thinking-2507-FP8 Everything that runs on NVIDIA B200

Files and context on NVIDIA B200

Each file of Qwen3-235B-A22B-Thinking-2507-FP8 (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.

FileSizeOn NVIDIA B200 Context (16-bit)Context (8-bit)
As published236 GBDoesn't fit——
4-bit (estimate)132 GBTight151K256K

Qwen3-235B-A22B-Thinking-2507-FP8 on other hardware

Popular models that run on NVIDIA B200

Assumes 90% of the card's memory is usable and at least 4K tokens of context for one conversation, with the context cache in 16-bit unless noted. Fits leaves at least 20% of that memory for context; Tight leaves less.