Skip to content

Can Radeon RX 7900 GRE run K2 Horizon 7B?

Comfortable: GGUF Q8_0 needs 10.6 GB at 4K, batch 1; headroom 5.4 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 Q4_K_M6.5ComfortableVerified file
GGUF Q5_K_M7.4ComfortableVerified file
GGUF Q6_K8.3ComfortableVerified file
GGUF Q8_010.6ComfortableVerified file

Context capacity · GGUF Q8_0

ContextTotal GBVerdict
4K10.6Comfortable
16K12.5Comfortable
32K14.9Tight
128K29.8Does not fit

Generation speed

GGUF Q8_0: 38–50 tok/s · Derived range

Other model / GPU pairs