Hardware

Radeon RX 7900 XT

Radeon RX 7900 XT has 20 GB GDDR6 and 800 GB/s. See which open models fit, at which quant, and roughly how fast. Figures below use the default memory (20 GB), 32 GB system RAM, and DDR5-5600 for anything that spills. Updated 2026-10-17.

Memory20 GB GDDR6

Bandwidth800 GB/s

Kindgpu

Year2022

20GB GDDR6, 320-bit, 800 GB/s. Source.

RSS for new models on this device

Models that fit

Model	Verdict	Quant	Context	tok/s
GLM 4 32B 0414 32.6B	Tight fit	Q4_K_S	4K	23–32
Qwen3.8 27B 27.3B	Tight fit	UD-Q4_K_XL	8K	24–33
gemma 3 27b it 27.0B	Tight fit	UD-Q4_K_XL	8K	21–29
medgemma 27b text it 27.0B	Tight fit	UD-Q4_K_XL	8K	21–29
Qwen3.6 27B 26.9B	Tight fit	UD-Q4_K_XL	8K	24–33
Qwen3.5 27B 26.9B	Tight fit	UD-Q4_K_XL	8K	24–33
Qwen2.5 32B Instruct 32.8B	Tight fit	IQ4_XS	4K	23–32
Qwen3 32B 32.8B	Tight fit	IQ4_XS	4K	23–32
Qwen2.5 Coder 32B Instruct 32.8B	Tight fit	IQ4_XS	4K	23–32
Qwen3 VL 32B Instruct 32.8B	Tight fit	IQ4_XS	4K	23–32
gemma 4 31B it 30.7B	Tight fit	Q4_K_S	4K	21–28
Devstral Small 2 24B Instruct 2512 23.6B	Tight fit	UD-Q5_K_XL	8K	24–33
Mistral Small 3.2 24B Instruct 2506 23.6B	Tight fit	UD-Q5_K_XL	8K	24–33
Magistral Small 2509 23.6B	Tight fit	UD-Q5_K_XL	8K	24–33
Llama 3.3 70B Instruct 70.6B	Tight fit	UD-IQ1_M	4K	24–32
DeepSeek R1 Distill Llama 70B 70.6B	Tight fit	UD-IQ1_M	4K	24–32
Hermes 4 70B 70.6B	Tight fit	UD-IQ1_M	4K	24–32
GLM 4.7 Flash 29.9B	Tight fit	UD-Q4_K_XL	8K	106–144
Qwen3 Coder 30B A3B Instruct 30.5B	Tight fit	UD-Q4_K_XL	8K	173–236
Qwen3 30B A3B Instruct 2507 30.5B	Tight fit	UD-Q4_K_XL	8K	173–236
Qwen3 VL 30B A3B Instruct 30.5B	Tight fit	UD-Q4_K_XL	8K	173–236
Qwen3 30B A3B Thinking 2507 30.5B	Tight fit	UD-Q4_K_XL	8K	173–236
Qwen3.5 35B A3B 34.7B	Tight fit	UD-IQ4_XS	8K	262–357
Qwen3.6 35B A3B 34.7B	Tight fit	UD-IQ4_XS	8K	258–352
Kwaipilot KAT Coder V2.5 Dev 34.7B	Tight fit	IQ4_XS	4K	126–172
DeepSeek R1 Distill Qwen 32B 32.8B	Tight fit	Q3_K_M	8K	24–33
Nemotron 3 Nano 30B A3B 31.6B	Tight fit	IQ4_NL	8K	203–277
LFM2.5 24B A2B 24.0B	Tight fit	UD-Q5_K_XL	8K	203–277
gemma 4 26B A4B it 25.2B	Tight fit	UD-Q4_K_XL	8K	93–127
GLM 4.7 Flash REAP 23B A3B 23.0B	Tight fit	UD-Q5_K_XL	8K	167–228
gpt oss 20b 20.9B	Runs	UD-Q8_K_XL	8K	109–149
Qwen3 14B 14.8B	Tight fit	Q8_0	8K	26–35
NousResearch Hermes 4 14B 14.8B	Tight fit	Q8_0	8K	26–35
Qwen2.5 14B Instruct 14.8B	Tight fit	Q8_0	8K	25–35
DeepSeek R1 Distill Qwen 14B 14.8B	Tight fit	Q8_0	8K	25–35
Qwen2.5 Coder 14B Instruct 128K 14.8B	Tight fit	Q8_0	8K	25–35
Qwen3.8 35B A3B Distill 35.5B	Tight fit	Q3_K_M	8K	265–361
phi 4 14.7B	Tight fit	Q8_0	8K	25–35
Ministral 3 14B Instruct 2512 13.5B	Tight fit	UD-Q8_K_XL	8K	24–33
Ministral 3 14B Reasoning 2512 13.5B	Tight fit	UD-Q8_K_XL	8K	24–33

90 more models in the catalog do not fit this default profile, or ranked below the first 40 that do.

Example: Tight fit fully on your Radeon RX 7900 XT at Q4_K_S, 4K context, about 23–32 tok/s. Weights 17.4 GB plus 0.24 GB of KV cache — 18.1 GB of 18.5 GB usable. Weights shown in the verdict use GiB (the same 1024³ counting as GPU memory).