Skip to content

Can A100 80G run Xing4.0 29B-A4B?

Comfortable: GGUF Q8_0 needs 34.9 GB at 4K, batch 1; headroom 45.1 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 Q8_034.9ComfortableVerified file
GGUF Q6_K28.6ComfortableVerified file
GGUF Q5_K_M25.3ComfortableVerified file
GGUF Q4_K_M21.2ComfortableVerified file
GGUF IQ4_XS19.0ComfortableVerified file
GGUF Q3_K_M17.0ComfortableVerified file
GGUF IQ3_XXS15.3ComfortableVerified file
GGUF Q2_K13.6ComfortableVerified file
GGUF IQ2_XXS11.0ComfortableVerified file

Context capacity · GGUF Q8_0

ContextTotal GBVerdict
4K34.9Comfortable
16K35.5Comfortable
32K36.3Comfortable
128K40.9Comfortable

Generation speed

GGUF Q8_0: 479 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs