Hardware / AMD /Radeon RX 7900 XTX

AMD · GPU

Radeon RX 7900 XTX

Runs 125 models fully in memory, 9 more with CPU offload. 9 of those sit close enough to 24 GB that our size estimate is what decides them.

Memory

24 GB

Bandwidth

960 GB/s

Runs

125

01

Specifications

Memory

24 GB

Decides which models fit at all.

Memory bandwidth

960 GB/s

How fast it can read the model. Sets the speed of the reply.

Compute

122.8 TFLOPS

Trillions of half-precision sums a second. Sets how fast it reads a long prompt, not how fast it replies.

Power draw

355 W

At full load, for sizing a power supply.

MSRP

$999

What the maker asked for it new. We hold no used or street prices, so this is not what one costs today.

Released

Dec 13, 2022

Architecture

RDNA 3

02

What it runs

Best quant per model. Text models first, then speech, then what does not fit.

What is quantisation? →

ModelParamsContextSpeedVerdict

Ministral 3 3B 25123.8B131K271 tok/sFits in memory gpt-oss-safeguard-20b21.5B131K243 tok/sFits in memory Granite 4.0 Micro3.2B66K184 tok/sFits in memory UI-TARS 7B8.3B66K124 tok/sFits in memory Llama 3.2 3B Instruct3.2B131K322 tok/sFits in memory Qwen3 8B8.2B66K126 tok/sFits in memory Gemma 3 4B4.3B131K204 tok/sFits in memory Llama 3.1 8B Instruct8B131K129 tok/sFits in memory Nemotron 3.5 Content Safety4.3B131K204 tok/sFits in memory Granite 4.1 8B8.8B66K117 tok/sFits in memory Ministral 3 8B 25128.9B66K116 tok/sFits in memory Qwen3 VL 8B Thinking8.8B66K117 tok/sFits in memory Qwen3 VL 8B Instruct8.8B66K117 tok/sFits in memory Granite 4.2 8B8.8B66K117 tok/sFits in memory Qwen3.5-9B9.7B66K106 tok/sFits in memory Gemma 3n 4B7.8B33K132 tok/sFits in memory Llama 3.2 1B Instruct1.2B33K858 tok/sFits in memory Qwen2.5 7B Instruct7.6B33K135 tok/sFits in memory Gemma 4 26B A4B25.8B16K219 tok/sFits in memory Reka Edge7.1B16K145 tok/sFits in memory Nemotron 3 Nano 30B A3B31.6B16K292 tok/sFits in memoryest Qwen3 30B A3B Thinking 250730.5B16K292 tok/sFits in memoryest Qwen3 Coder 30B A3B Instruct30.5B16K292 tok/sFits in memoryest Qwen3 30B A3B Instruct 250730.5B16K292 tok/sFits in memoryest Qwen3 VL 30B A3B Thinking31.1B8K292 tok/sFits in memoryest Llama Guard 4 12B12B131K86 tok/sFits in memory Gemma 3 12B12.2B131K84 tok/sFits in memory Rocinante 12B12.2B66K84 tok/sFits in memory Llama 3 8B Lunaris8B8K129 tok/sFits in memory Mistral Nemo12.2B66K84 tok/sFits in memory Qwen3 VL 30B A3B Instruct31.1B8K292 tok/sFits in memoryest Hy-MT2-1.8B2B8K437 tok/sFits in memory Hy-MT2-30B-A3B30.1B8K292 tok/sFits in memoryest Ministral 3 14B 251213.9B66K74 tok/sFits in memory UnslopNemo 12B12.2B33K84 tok/sFits in memory Hy-MT2-7B8B8K109 tok/sFits in memory Qwen3 14B14.8B66K70 tok/sFits in memory Phi 414.7B16K70 tok/sFits in memory MythoMax 13B13B8K79 tok/sFits in memory Reka Flash 320.9B33K49 tok/sFits in memory Cydonia 24B V4.123.6B33K44 tok/sFits in memory Uncensored24B33K43 tok/sFits in memory Mistral Small 3.2 24B24B33K43 tok/sFits in memory ReMM SLERP 13B13B4K79 tok/sFits in memory Mistral Small 3.1 24B24B33K43 tok/sFits in memory Mistral Small 323.6B33K44 tok/sFits in memory Ternary Bonsai 2 27B27B33K38 tok/sFits in memory Gemma 3 27B27.4B33K38 tok/sFits in memory Muse Glimmer 30B29.8B33K35 tok/sFits in memory Voxtral Small 24B 250724.3B16K42 tok/sFits in memory Nemotron 3.5 Lightning31.6B16K28 tok/sFits in memoryest Qwen3.5-27B27.8B8K37 tok/sFits in memory Qwen3.6 27B27.8B8K37 tok/sFits in memory Gemma 2 27B27.2B8K38 tok/sFits in memory Qwen3.8 27B27.8B8K37 tok/sFits in memory GLM 4.7 Flash31.2B4K28 tok/sFits in memory Laguna XS 2.133.4B4K26 tok/sFits in memory Olmo 3 32B Think32.2B2K32 tok/sFits in memoryest

Speech models it runs67 modelsTranscription and speech models that fit in memory. Throughput figures for a transcription model come from its text decoder, so treat this as a fit answer rather than a speed one.

Niagara 19m Batch.en20M29365 tok/sFits in memory Niagara 38m Batch.en38M15455 tok/sFits in memory Higgs Audio v3 STT2.7B218 tok/sFits in memory Cohere Transcribe 03 20262.1B280 tok/sFits in memory Distil Large v3.50.8B734 tok/sFits in memory Lite Whisper Large v3 Acc1.4B627 tok/sFits in memory Owsm CTC v3.1 1B1.1B936 tok/sFits in memory Owsm CTC v3.2 ft 1B1B587 tok/sFits in memory Owsm CTC v4 1B1B1029 tok/sFits in memory Hubert Large Ls960 ft0.3B1835 tok/sFits in memory MMS 1b All1B877 tok/sFits in memory Wav2vec2 Large 960h Lv60 Self0.3B1835 tok/sFits in memory Hojo ASR V15.2B169 tok/sFits in memory Granite 4.0 1b Speech2.3B255 tok/sFits in memory Granite Speech 3.3 2b3B343 tok/sFits in memory Granite Speech 3.3 8b8.6B120 tok/sFits in memory Granite Speech 4.1 2b2.3B448 tok/sFits in memory Granite Speech 4.1 2b NAR2.3B448 tok/sFits in memory STT 2.6b en2.6B396 tok/sFits in memory Phi 4 Multimodal Instruct5.6B131K184 tok/sFits in memory VibeVoice ASR HF8.3B124 tok/sFits in memory Voxtral Mini 3B 25074.7B125 tok/sFits in memory Voxtral Mini 4B Realtime 26024.4B134 tok/sFits in memory Canary 180m Flash0.2B4874 tok/sFits in memory Canary 1b1B587 tok/sFits in memory Canary 1b Flash0.8B734 tok/sFits in memory Canary 1b v21B587 tok/sFits in memory Canary Qwen 2.5b2.6B226 tok/sFits in memory Nemotron 3.5 ASR Streaming 0.6b0.6B979 tok/sFits in memory Nemotron Speech Streaming en 0.6b0.6B979 tok/sFits in memory Parakeet CTC 0.6b0.6B1716 tok/sFits in memory Parakeet CTC 1.1b1.1B534 tok/sFits in memory Parakeet RNNT 0.6b0.6B979 tok/sFits in memory Parakeet RNNT 1.1b1.1B936 tok/sFits in memory Parakeet TDT CTC 110m0.1B5339 tok/sFits in memory Parakeet TDT 0.6b v20.6B1462 tok/sFits in memory Parakeet TDT 0.6b v30.6B979 tok/sFits in memory Parakeet TDT 1.1b1.1B534 tok/sFits in memory Nyra Health CrisperWhisper1.6B367 tok/sFits in memory Whisper Large v31.5B585 tok/sFits in memory Whisper Large v3 Turbo0.8B1287 tok/sFits in memory MOSS Transcribe Diarize0.9B653 tok/sFits in memory Qwen3 ASR 0.6B0.9B1144 tok/sFits in memory Qwen3 ASR 0.6B HF0.8B1287 tok/sFits in memory Qwen3 ASR 1.7B2.3B381 tok/sFits in memory Qwen3 ASR 1.7B HF2B294 tok/sFits in memory Zipformer cr CTC Transducer XL 290M0.3B3025 tok/sFits in memory Zipformer Transducer XL 290M0.3B2025 tok/sFits in memory ASR Conformer Largescaleasr0.5B2144 tok/sFits in memory GLM ASR Nano 25122.3B381 tok/sFits in memory Granite Speech 5.0 470m Turboctc0.5B1755 tok/sFits in memory Granite Speech 5.0 470m Turboctc nc0.5B2059 tok/sFits in memory Fun ASR Nano 2512 HF0.8B1287 tok/sFits in memory Gemma 4 E2B it5.1B202 tok/sFits in memory Gemma 4 E4B it8B131K129 tok/sFits in memory Niagara 9m Batch.en9M65255 tok/sFits in memory Niagara 84m Batch.en84M6992 tok/sFits in memory Higgs Audio v3 8b STT v28.9B116 tok/sFits in memory ARK ASR 0.6B1.3B33K675 tok/sFits in memory ARK ASR 3B4.1B33K251 tok/sFits in memory Audio8 ASR 0.1B0.3B33K1958 tok/sFits in memory MOSS Transcribe Preview 2B2.4B33K429 tok/sFits in memory Moonshine Streaming Medium0.3B4K3431 tok/sFits in memory Moonshine Streaming Small0.1B4K5873 tok/sFits in memory Gemma 4 12B it12B86 tok/sFits in memory Moonshine Streaming Tiny0B4Kno speed estimateFits in memory Moonshine Tiny0B194no speed estimateFits in memory

Runs only by spilling into system memory9 modelsThese load, but part of the weights sits in ordinary system memory, which is far slower than the chip.

Nex-N2-Mini35.1B—no speed estimateSpills to system RAMest Qwen3.6 35B A3B36B—no speed estimateSpills to system RAMest Gemma 4 31B31.3B—no speed estimateSpills to system RAM Qwen3 VL 32B Instruct33.4B—no speed estimateSpills to system RAM Skyfall 36B V236.9B—no speed estimateSpills to system RAM Qwen2.5 Coder 32B Instruct32.8B—no speed estimateSpills to system RAMest Qwen3.5-35B-A3B36B—no speed estimateSpills to system RAMest Qwen3 32B32.8B—no speed estimateSpills to system RAMest Nex-N2.5-Mini35.1B—no speed estimateSpills to system RAM

Too large for this device90 modelsThe weights do not fit even with part of them offloaded to system memory. Closest calls first.

Llama 3.1 70B Instruct70.6B—no speed estimateToo large Llama 3.3 Euryale 70B70.6B—no speed estimateToo large Qwen3 Next 80B A3B Thinking81.3B—no speed estimateToo large Qwen3 Coder Next79.7B—no speed estimateToo large Llama 3.3 70B Instruct70.6B—no speed estimateToo large Qwen2.5 72B Instruct72.7B—no speed estimateToo large GLM 4.5V108B—no speed estimateToo large Command A111B—no speed estimateToo large gpt-oss-120b120B—no speed estimateToo large Hermes 4 70B70.6B—no speed estimateToo large Llama 3.1 Euryale 70B v2.270.6B—no speed estimateToo large R1 Distill Llama 70B70.6B—no speed estimateToo large Hermes 3 70B Instruct70.6B—no speed estimateToo large GLM 4.6V108B—no speed estimateToo large Ling-2.6-flash107B—no speed estimateToo large Magnum v4 72B72.7B—no speed estimateToo large Mistral Medium 3.5128B—no speed estimateToo large Qwen2.5 VL 72B Instruct73.4B—no speed estimateToo large Laguna S 2.1118B—no speed estimateToo large Hunyuan A13B Instruct80.4B—no speed estimateToo large Mistral Small 4119B—no speed estimateToo large Qwen3 Next 80B A3B Instruct81.3B—no speed estimateToo large Nemotron 3 Super124B—no speed estimateToo large Qwen3.5-122B-A10B125B—no speed estimateToo large Devstral 2 2512125B—no speed estimateToo large Mixtral 8x22B Instruct141B—no speed estimateToo large Llama 4 Scout109B—no speed estimateToo large GLM 4.5 Air111B—no speed estimateToo large Step 3.7 Flash201B—no speed estimateToo large Ling-3.0-flash128B—no speed estimateToo large MiniMax M2229B—no speed estimateToo large Step 3.5 Flash199B—no speed estimateToo large Qwen3 VL 235B A22B Instruct236B—no speed estimateToo large MiniMax M2.5229B—no speed estimateToo large Qwen3 235B A22B Instruct 2507235B—no speed estimateToo large Qwen3 VL 235B A22B Thinking236B—no speed estimateToo large DeepSeek V4 Flash291B—no speed estimateToo large GLM 5.3 Flash321B—no speed estimateToo large Hy3299B—no speed estimateToo large Laguna M.1226B—no speed estimateToo large Nex-N2-Pro397B—no speed estimateToo large Trinity Large Thinking399B—no speed estimateToo large MiniMax M2.7229B—no speed estimateToo large MiniMax M2.1229B—no speed estimateToo large Qwen3 235B A22B Thinking 2507235B—no speed estimateToo large GLM 4.6357B—no speed estimateToo large GLM 4.7358B—no speed estimateToo large ERNIE 4.5 VL 424B A47B424B—no speed estimateToo large MiniMax M3427B—no speed estimateToo large Nex-N2.5-Pro397B—no speed estimateToo large Inkling Small266B—no speed estimateToo large Jamba Large 1.7399B—no speed estimateToo large Llama 4 Maverick402B—no speed estimateToo large Qwen3 Coder 480B A35B480B—no speed estimateToo large Hy3 preview299B—no speed estimateToo large DeepSeek V4 Flash Vision Exp305B—no speed estimateToo large MiniMax-01456B—no speed estimateToo large MiMo-V2.5311B—no speed estimateToo large GLM 4.5358B—no speed estimateToo large DeepSeek V3 0324685B—no speed estimateToo large Qwen3.5 397B A17B403B—no speed estimateToo large Hermes 4 405B406B—no speed estimateToo large Hermes 3 405B Instruct406B—no speed estimateToo large GLM 5.1754B—no speed estimateToo large MiniMax M1456B—no speed estimateToo large DeepSeek V3.1 Terminus685B—no speed estimateToo large DeepSeek V3685B—no speed estimateToo large GLM 5754B—no speed estimateToo large DeepSeek V4.1 Flash763B—no speed estimateToo large Inkling952B—no speed estimateToo large Nemotron 3 Ultra561B—no speed estimateToo large MiMo-V2.5-Pro1T—no speed estimateToo large Kimi K2 07111T—no speed estimateToo large Ling-2.6-1T1T—no speed estimateToo large Kimi K2.7 Code1.1T—no speed estimateToo large R1 0528685B—no speed estimateToo large DeepSeek V3.1685B—no speed estimateToo large DeepSeek V3.2685B—no speed estimateToo large DeepSeek V3.2 Exp685B—no speed estimateToo large Kimi K2 09051T—no speed estimateToo large Ring-2.6-1T1T—no speed estimateToo large Kimi K2.61.1T—no speed estimateToo large Kimi K2.51.1T—no speed estimateToo large GLM 5.2753B—no speed estimateToo large Hy4 preview780B—no speed estimateToo large Kimi K2 Thinking1.1T—no speed estimateToo large Qwen3.8 2.4T A95B2.4T—no speed estimateToo large DeepSeek V4 Pro1.6T—no speed estimateToo large LongCat 2.01.8T—no speed estimateToo large Kimi K32.8T—no speed estimateToo large

224 models have a verdict on this device: 125 fit in memory, 9 spill into system memory and 90 are too large. Every one of them is on this page.

03

Other RDNA 3 cards

The same generation in other memory sizes. Memory is what decides which models fit.

Radeon RX 7900 XT20 GB · less memory

Something wrong on this page? Tell us