Consumer · GDDR6
Radeon RX 7900 XTX 24 GB: which LLMs can it run?
On its own, it loads up to DeepSeek R1 Distill 32B in Q4. With several cards, memory adds up and opens bigger models.
Specifications
Memory24 GB
Bandwidth960 GB/s
FP16 compute~123 TF
Power355 W
Indicative price€950
Cloud rental€0.30/h
Against cards at the same price
Models and expected speed
| Model | Q4 | Q8 | FP16 |
|---|---|---|---|
| Qwen3 4B | ✓ ~207 tok/s | ✓ ~120 tok/s | ✓ ~65 tok/s |
| Llama 3.1 8B | ✓ ~107 tok/s | ✓ ~61 tok/s | ✓ ~33 tok/s |
| Qwen3 8B | ✓ ~104 tok/s | ✓ ~60 tok/s | ✓ ~32 tok/s |
| Gemma 3 12B | ✓ ~72 tok/s | ✓ ~41 tok/s | 2 × · ~33 tok/s |
| Mistral Nemo 12B | ✓ ~71 tok/s | ✓ ~40 tok/s | 2 × · ~32 tok/s |
| Phi-4 14B | ✓ ~59 tok/s | ✓ ~34 tok/s | 2 × · ~27 tok/s |
| Qwen3 14B | ✓ ~58 tok/s | ✓ ~33 tok/s | 2 × · ~27 tok/s |
| gpt-oss-20b | ✓ ~229 tok/s | 2 × · ~200 tok/s | 2 × · ~108 tok/s |
| Mistral Small 24B | ✓ ~36 tok/s | 2 × · ~31 tok/s | 3 × · ~25 tok/s |
| Gemma 3 27B | ✓ ~32 tok/s | 2 × · ~28 tok/s | 3 × · ~22 tok/s |
| Qwen3.8 27B | ✓ ~31 tok/s | 2 × · ~27 tok/s | 3 × · ~21 tok/s |
| Qwen3 30B-A3B | ✓ ~248 tok/s | 2 × · ~217 tok/s | 3 × · ~176 tok/s |
| Qwen3 32B | ✓ ~27 tok/s | 2 × · ~23 tok/s | 4 × · ~24 tok/s |
| Qwen2.5 Coder 32B | ✓ ~27 tok/s | 2 × · ~23 tok/s | 4 × · ~24 tok/s |
| DeepSeek R1 Distill 32B | ✓ ~27 tok/s | 2 × · ~23 tok/s | 4 × · ~24 tok/s |
| Mixtral 8x7B | 2 × · ~100 tok/s | 3 × · ~86 tok/s | ✗ |
| Llama 3.3 70B | 3 × · ~28 tok/s | 4 × · ~21 tok/s | ✗ |
| GLM-4.5-Air 106B | 3 × · ~162 tok/s | ✗ | ✗ |
| gpt-oss-120b | 4 × · ~493 tok/s | ✗ | ✗ |
| Mistral Large 123B | 4 × · ~21 tok/s | ✗ | ✗ |
| Qwen3 235B-A22B | ✗ | ✗ | ✗ |
| GLM-4.5 355B | ✗ | ✗ | ✗ |
| Llama 3.1 405B | ✗ | ✗ | ✗ |
| DeepSeek R1 671B | ✗ | ✗ | ✗ |
| DeepSeek V3.1 671B | ✗ | ✗ | ✗ |
| Kimi K2 1T | ✗ | ✗ | ✗ |
8k tokens, one user. “2 ×” = number of cards needed, speed with tensor parallelism (vLLM). ✗ = doesn’t fit, even with 4 cards.