24 GB · 1,008 GB/s
What can the NVIDIA RTX 4090 run?
Still a top 24 GB card. Runs the best 27–32B models at a smooth speed.
Best models for each task
Our top three for this machine. Chat and image tasks assume an 8K-token conversation; coding, documents and agents assume 32K.
Chat & writing18 of 25 run
- Qwen3.8 27BRuns wellQ4_K_M · needs 19.9 GB at 8K · 35–50tok/sFast
- Gemma 4 31BRuns wellQ4_K_M · needs 21.4 GB at 8K · 30–50tok/sFast
- Muse Glimmer 30BRuns wellQ4_K_M · needs 19.8 GB at 8K · 35–55tok/sFast
Coding20 of 26 run
- Qwen3.8 27BRuns wellQ4_K_M · needs 21.5 GB at 32K · 30–50tok/sSmooth
- Muse Glimmer 30BRuns wellQ4_K_M · needs 20.2 GB at 32K · 35–50tok/sFast
- Qwen3.6 35B-A3BRuns wellUD-IQ4_XS · needs 20.3 GB at 32K · 110–210tok/sFast
Long documents13 of 20 run
- Qwen3.8 27BRuns wellQ4_K_M · needs 21.5 GB at 32K · 30–50tok/sSmooth
- Qwen3.6 35B-A3BRuns wellUD-IQ4_XS · needs 20.3 GB at 32K · 110–210tok/sFast
- Gemma 4 12BRuns greatQ8_0 · needs 14.7 GB at 32K · 45–70tok/sFast
Agents & tools20 of 26 run
- Qwen3.8 27BRuns wellQ4_K_M · needs 21.5 GB at 32K · 30–50tok/sSmooth
- Qwen3.6 35B-A3BRuns wellUD-IQ4_XS · needs 20.3 GB at 32K · 110–210tok/sFast
- Muse Glimmer 30BRuns wellQ4_K_M · needs 20.2 GB at 32K · 35–50tok/sFast
Images in9 of 13 run
- Qwen3.8 27BRuns wellQ4_K_M · needs 19.9 GB at 8K · 35–50tok/sFast
- Qwen3.6 35B-A3BRuns wellUD-IQ4_XS · needs 19.8 GB at 8K · 120–240tok/sFast
- Muse Glimmer 30BRuns wellQ4_K_M · needs 19.8 GB at 8K · 35–55tok/sFast
Make images6 of 6 run
- Stable Diffusion XL Base 1.0Runs greatFP16 · needs 7.9 GB
- FLUX.2 [klein] 4BRuns greatBF16 · needs 14.0 GB
- Z-Image-TurboRuns greatBF16 · needs 17.0 GB
Speak text7 of 7 run
- Qwen3-TTS 1.7BRuns greatBF16 · needs 5.5 GB
- Kokoro-82MRuns greatFP32 · needs 1.4 GB
- Chatterbox Multilingual V3Runs greatFull · needs 4.2 GB
Transcribe7 of 7 run
- Voxtral Mini 4B RealtimeRuns greatBF16 · needs 17.0 GB
- Whisper large-v3-turbo (via faster-whisper)Runs greatFP16 · needs 2.6 GB
- Nemotron 3.5 ASR Streaming 0.6BRuns greatFP32 · needs 3.5 GB
Biggest model it runs well
Qwen3.6 35B-A3B (UD-IQ4_XS, 17.7 GB file) at about 120–240tok/sFast.
Should you buy it?
Buy it if
You find a used one from a trusted seller at a fair price.
Skip it if
A used RTX 3090 has the same 24 GB for much less, if a little slower.
Watch out for
- Large card (3–3.5 slots); check your case.
- Needs 850 W+ power supply.
Measured on this hardware
Real runs from public sources, not estimates. All measurements.