Skip to content

Can Tesla P40 24G run Xing4.0 29B-A4B?

Comfortable: GGUF IQ4_XS needs 19.0 GB at 4K, batch 1; headroom 5.0 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_034.9Does not fitVerified file
GGUF Q6_K28.6Does not fitVerified file
GGUF Q5_K_M25.3Does not fitVerified file
GGUF Q4_K_M21.2TightVerified 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 IQ4_XS

ContextTotal GBVerdict
4K19.0Comfortable
16K19.6Comfortable
32K20.4Comfortable
128K25.0Tight

Generation speed

GGUF IQ4_XS: 158 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs