Comment by fluidcruft

3 days ago

idk but naive me sees the article say the model needed 158GB of RAM so I guess 10x 16GB GPUs? Apparently retail price is $1k per 5080 16GB but they're actually selling at $1.5k so I guess $15k for just the GPUs.

4 comments

fluidcruft

Reply
petu  3 days ago

158GB number is for Mistral Medium 3.5 128B model, Muse Spark is way larger.

  Muse Spark Open would require more than 1.6TB of RAM in its native precision (BF16) for the model parameters alone. In a 4-bit quantized (NVFP4) format, the model memory footprint is lowered to about 493GB.

  • bigyabai  3 days ago

    So about 24gb/expert at NVFP4. If you compressed the KV cache, you might be able to get away with 2x 24GB cards and stream the experts.