RX 7700 XT LLM benchmarks

18 results · 70 measurements · 2 sources

Explore all results Image benchmarks Submit a result

Device specifications

Variant	Memory	Memory type	Bus	Bandwidth	FP16 matrix	INT8 matrix
RX 7700 XT ↗	12 GB	GDDR6	192-bit	432 GB/s	70.3 TFLOPS	70.3 TOPS

Manufacturer specifications. Shared system RAM is not dedicated VRAM. Unreported values remain —; measured LLM results retain their reported configuration.

Selected reported tests

18 of 18 results

Source-reported results, selected across reports by documentation quality. Match configurations before comparing values.


View ReportLocalMaxxing	Laguna-XS-2.1Q4_K_SWeights ↗	Token speed	—	246.3 tok/sPrefill · unknown	13.7 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 63 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-23 Full record →
View ReportGeneric LLM	gemma-4-12B-itQAT-Q4_0Model ref. ↗	Token speed	—	~500 tok/sPrefill · range lower · per request ~900 tok/sPrefill · range upper · per request	~45 tok/sGeneration · per request	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine llama.cpp Backend ROCm / HIP Harness Author daily-use timing Prompt / output ? / ? tokens Context 65536 Concurrency Not reported Date Not reported Full record →
View ReportLocalMaxxing	Qwythos-9B-Claude-Mythos-5-1MQ4_K_MWeights ↗	Token speed	—	314.3 tok/sPrefill · unknown	53.6 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 62 / 669 tokens Context 2048 Concurrency Not reported Date 2026-07-15 Full record →
View ReportGeneric LLM	Qwen2.5-Coder-14B-InstructQ4_0Model ref. ↗	Token speed	—	~400 tok/sPrefill · range lower · per request ~800 tok/sPrefill · range upper · per request	~35 tok/sGeneration · range lower · per request ~38 tok/sGeneration · range upper · per request	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine llama.cpp Backend ROCm / HIP Harness Author daily-use timing Prompt / output ? / ? tokens Context 32768 Concurrency Not reported Date Not reported Full record →
View ReportLocalMaxxing	Qwythos-9B-v2Q4_K_MWeights ↗	Token speed	—	172.6 tok/sPrefill · unknown	49.3 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 62 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportGeneric LLM	Qwen2.5-Coder-14B-InstructQ4_0Model ref. ↗	Token speed	—	~400 tok/sPrefill · range lower · per request ~700 tok/sPrefill · range upper · per request	~35 tok/sGeneration · per request	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine llama.cpp Backend ROCm / HIP Harness Author daily-use timing Prompt / output ? / ? tokens Context 32768 Concurrency Not reported Date Not reported Full record →
View ReportLocalMaxxing	Qwythos-9B-v2Q4_K_MWeights ↗	Token speed	—	246.9 tok/sPrefill · unknown	50.2 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 62 / 256 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportLocalMaxxing	gemma-4-12B-it-qatQ4_0Weights ↗	Token speed	—	163.9 tok/sPrefill · unknown	40.4 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 37 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-25 Full record →
View ReportLocalMaxxing	gemma-4-E4BQ4_K_MModel ref. ↗	Token speed	—	313.6 tok/sPrefill · unknown	71.2 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 37 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-25 Full record →
View ReportLocalMaxxing	Agents-A1-4BQ4_K_MWeights ↗	Token speed	—	2,442.9 tok/sPrefill · unknown	79 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 273 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-21 Full record →
View ReportLocalMaxxing	Agents-A1-4BQ4_K_MWeights ↗	Token speed	—	3,135.6 tok/sPrefill · unknown	76.3 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 273 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-21 Full record →
View ReportLocalMaxxing	Grug-12BQ4_K_MWeights ↗	Token speed	—	81.8 tok/sPrefill · unknown	15.5 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 49 / 256 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportLocalMaxxing	Ornith-1.0-9BQ5_K_MModel ref. ↗	Token speed	—	369.4 tok/sPrefill · unknown	30.6 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 32 / 256 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportLocalMaxxing	Bonsai-27BQ1_0Weights ↗	Token speed	—	272.5 tok/sPrefill · unknown	44 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 32 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportLocalMaxxing	LFM2.5-8B-A1BIQ4_XSModel ref. ↗	Token speed	—	478.4 tok/sPrefill · unknown	156.7 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 31 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-16 Full record →
View ReportLocalMaxxing	LFM2.5-8B-A1BIQ4_XSWeights ↗	Token speed	—	132.9 tok/sPrefill · unknown	150.8 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 31 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportLocalMaxxing	LFM2.5-8B-A1BIQ4_XSWeights ↗	Token speed	—	764.1 tok/sPrefill · unknown	153.9 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 31 / 256 tokens Context 2048 Concurrency Not reported Date 2026-07-14 Full record →
View ReportLocalMaxxing	GLM-4.7-FlashQ4_K_MModel ref. ↗	Token speed	—	264.4 tok/sPrefill · unknown	21.5 tok/sGeneration · unknown	1 × RX 7700 XTSingle GPU · 12 GB VRAM/GPU Configuration Memory 12 GB VRAM/GPU OS Windows 11 Engine vllm Backend Not reported Harness LocalMaxxing community speed test; exact client where reported Prompt / output 27 / 1024 tokens Context 2048 Concurrency Not reported Date 2026-07-16 Full record →

Explore all 18 results →

Models and software in this collection

Model families: GLM 4.7, Gemma 4, Laguna, Qwen2.5

Engines: llama.cpp, vllm

Methodology and evidence standards · Download the research dataset

RX 7700 XT LLM benchmarks

Selected reported tests

LLMs benchmarked on the RX 7700 XT