RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

Can Flint run on the Ryzen AI Max+ 395 128GB?

Yes. At 8K context the best fit is F16, using about 75.1 GB of the 109 GB usable on the Ryzen AI Max+ 395 128GB. Estimated speed ceiling: ~102 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •15.2 GB Fits15.4 GB Fits16.0 GB Fits18.6 GB Fits526
3-bit (Q3_K_M) •19.0 GB Fits19.1 GB Fits19.8 GB Fits22.4 GB Fits417
4-bit (Q4_K_M) •24.4 GB Fits24.5 GB Fits25.2 GB Fits27.8 GB Fits321
5-bit (Q5_K_M) •29.5 GB Fits29.6 GB Fits30.3 GB Fits32.9 GB Fits264
6-bit (Q6_K) •33.6 GB Fits33.7 GB Fits34.4 GB Fits37.0 GB Fits231
8-bit (Q8_0) •40.3 GB Fits40.4 GB Fits41.0 GB Fits43.7 GB Fits192
16-bit (F16/BF16) •75.0 GB Fits75.1 GB Fits75.8 GB Fits78.4 GB Fits102

• real GGUF file size from bartowski/RecursiveSI_flint-GGUF. Speed is a bandwidth ceiling (256 GB/s), not a benchmark. Methodology.

What fits instead

"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.

More

Flint on every GPU and Mac · Every model that runs on the Ryzen AI Max+ 395 128GB · Other models that fit 64 GB VRAM · Try other settings in the calculator

Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs

Data: Hugging Face Hub (RecursiveSI/flint), AMD specs. Model data last checked 2026-11-18; refreshed hourly.