RX 9070 16GB vs RX 9070 XT 16GB for Local LLMs

Both list the same memory bandwidth (640 GB/s), so per-token decode speed should be similar whenever a model fits both. Both fit the same number of catalog models at Q4/8K (25 of 61; 16 vs 16 GB VRAM). MSRP is $549 for the RX 9070 16GB vs $599 for the RX 9070 XT 16GB. MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. In the Q4/8K head-to-head below (44 shared models), the RX 9070 XT 16GB posts the higher tok/s in 2 rows vs 1. Methodology.

Spec comparison

	RX 9070 16GB	RX 9070 XT 16GB
Vendor	AMD	AMD
VRAM	16 GB	16 GB
Memory type	GDDR6	GDDR6
Memory bandwidth	640 GB/s	640 GB/s
TDP	220 W	304 W
MSRP (list)	$549	$599
Buy	Search Amazon (affiliate)	Search Amazon (affiliate)

MSRP is the launch list price, not live retail. Buy links are Amazon affiliate links — the price after you click is the live price.

Q4 / 8K head-to-head — top 15 of 44 shared models

Model	Params (B)	RX 9070 16GB tok/s	RX 9070 16GB fits	RX 9070 XT 16GB tok/s	RX 9070 XT 16GB fits	Winner
Qwen3-0.6B	0.8	509.1 est.	Yes	509.1 est.	Yes	Tie
gemma-3-1b-it	1.0	407.3 est.	Yes	407.3 est.	Yes	Tie
Llama-3.2-1B-Instruct	1.2	339.4 est.	Yes	339.4 est.	Yes	Tie
Qwen3-30B-A3B	30.5	293.9 est.	No	293.9 est.	No	Tie
Qwen3-Coder-30B-A3B-Instruct	30.5	293.9 est.	No	293.9 est.	No	Tie
Qwen3-30B-A3B-Instruct-2507	30.5	293.9 est.	No	293.9 est.	No	Tie
GLM-4.7-Flash	31.2	282.1 est.	No	282.1 est.	No	Tie
Qwen3-1.7B	2.0	203.6 est.	Yes	203.6 est.	Yes	Tie
Llama-3.2-3B-Instruct	3.2	127.3 est.	Yes	89.6 measured	Yes	RX 9070 16GB
gpt-oss-20b	21.5	113.1 est.	Yes	113.1 est.	Yes	Tie
Phi-4-mini-instruct	3.8	107.2 est.	Yes	107.2 est.	Yes	Tie
Phi-4-mini-reasoning	3.8	107.2 est.	Yes	107.2 est.	Yes	Tie
gemma-3-4b-it	4.3	104.4 est.	Yes	104.4 est.	Yes	Tie
Mistral-7B-Instruct-v0.3	7.2	101.3 measured	Yes	103.3 measured	Yes	RX 9070 XT 16GB
Qwen3-4B-Instruct-2507	4.0	101.8 est.	Yes	101.8 est.	Yes	Tie

est. = bandwidth-based estimate; measured = curated real benchmark that overrides the estimate. Fits = weights + KV(8K) + 1 GB overhead ≤ 95% of VRAM.

Fit coverage at Q4 / 8K

	RX 9070 16GB	RX 9070 XT 16GB
Catalog models that fit	25 of 61	25 of 61
Largest exclusive fit	—	—

RX 9070 16GB details → · RX 9070 XT 16GB details →