Can RTX 4090 24GB run gemma-4-31B-it?

NVIDIA · 24 GB GDDR6X · 1008 GB/s bandwidth · Google · 30.7B · model ctx up to 256K

Partly — gemma-4-31B-it at Q4 fits the RTX 4090 24GB (24 GB GDDR6X) at 4K context (≈ 20.96 GB) but not at the default 8K (≈ 24.03 GB needed). Use a shorter context or partial offload. Decode at Q4/8K: ≈ 59.5 tok/s (estimated, batch 1). FP16 needs ≈ 68.54 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	20.96 GB	Yes	20.9 est.
Q4	8K default	24.03 GB	No	59.5 est.
Q4	32K	42.45 GB	No	19.1 est.
FP16	4K	65.47 GB	No	2.2 est.
FP16	8K default	68.54 GB	No	2.0 est.
FP16	32K	86.96 GB	No	1.3 est.

Fits = weights + KV(ctx) + 1 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

Quant	Weights	KV cache	Overhead	Total needed	RTX 4090 24GB VRAM
Q4	16.89 GB	6.14 GB	1 GB	24.03 GB	24 GB
FP16	61.4 GB	6.14 GB	1 GB	68.54 GB	24 GB

Other GPUs that run gemma-4-31B-it

• B200 192GB SXM — ≈165.8 tok/s Q4 est.
• MI300X 192GB — ≈109.9 tok/s Q4 est.
• H200 141GB SXM5 — ≈99.5 tok/s Q4 est.
• H100 80GB SXM5 — ≈69.4 tok/s Q4 est.
• Gaudi 3 128GB — ≈67.9 tok/s Q4 est.
• A100 80GB SXM4 — ≈42.3 tok/s Q4 est.
• RTX 5090 32GB — ≈37.1 tok/s Q4 est.
• RTX 6000 Ada 48GB — ≈19.9 tok/s Q4 est.

All GPUs for gemma-4-31B-it →

Other models for the RTX 4090 24GB

• MiniCPM5-1B — ≈593.9 tok/s Q4 est.
• Qwen3.5-2B — ≈320.7 tok/s Q4 est.
• gpt-oss-20b — ≈225.2 tok/s Q4 measured
• GLM-4.7-Flash — ≈213.8 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈213.8 tok/s Q4 est.
• Qwen3.5-35B-A3B — ≈213.8 tok/s Q4 est.
• Qwen3.6-35B-A3B — ≈213.8 tok/s Q4 est.
• Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 — ≈213.8 tok/s Q4 est.

All models on the RTX 4090 24GB →