Skip to content

Can DGX Spark 128G run Recursive Flint 40B-A2.5B?

Comfortable: GGUF Q8_0 needs 44.5 GB at 4K, batch 1; headroom 83.5 GB.

Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.

Measured runs

No measured run for this pair. Sizing is an estimate, not a runtime test.

Quant memory at 4K

QuantTotal GBVerdictWeight sizing
GGUF Q3_K_M20.6ComfortableVerified file
GGUF Q4_K_M26.7ComfortableVerified file
GGUF Q8_044.5ComfortableVerified file

Context capacity · GGUF Q8_0

ContextTotal GBVerdict
4K44.5Comfortable
16K44.8Comfortable
32K45.3Comfortable
128K48.2Comfortable

Generation speed

GGUF Q8_0: 108 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs