RTX 4090 24GB for MiniMax H3: Speed, VRAM & Recommended Setup

24GB GDDR6X · bandwidth 1008 GB/s · Ada · sm_89 · estimated with 32GB RAM / fast-disk · updated 2026-09-28

Verdict: Recommended: 416 × 256 · 10s, ~1 min 11 s per clip (VRAM peak ~23GB). See the table for higher tiers.

Size (16:9)	5s	8s	10s	15s
416 × 256 0.1MP · flicker-prone	1 min 6 s VRAM 23G	1 min 9 s VRAM 23G	1 min 11 s VRAM 23G	⚠ 1 min 17 s VRAM 23G
864 × 480 0.4MP	⚠ 1 min 27 s VRAM 23G	⚠ 2 min VRAM 23G	⚠ 2 min 28 s VRAM 23G	⚠ 3 min 57 s VRAM 23G
960 × 544 0.5MP	⚠ 1 min 41 s VRAM 23G	⚠ 2 min 29 s VRAM 23G	⚠ 3 min 10 s VRAM 23G	⚠ 5 min 26 s VRAM 23G
1152 × 640 0.7MP	⚠ 2 min 13 s VRAM 23G	⚠ 3 min 41 s VRAM 23G	⚠ 4 min 59 s VRAM 23G	⚠ 9 min 27 s VRAM 23G
1344 × 768 1.0MP · official max	⚠ 3 min 10 s VRAM 23G	⚠ 5 min 54 s VRAM 23G	⚠ 8 min 26 s VRAM 23G	⚠ 17 min 43 s VRAM 23G
1376 × 768 1.0MP · above native	⚠ 3 min 15 s VRAM 23G	⚠ 6 min 7 s VRAM 23G	⚠ 8 min 46 s VRAM 23G	⚠ 18 min 35 s VRAM 23G
1920 × 1088 2.0MP · above native	⚠ 8 min 47 s VRAM 23G	⚠ 20 min 57 s VRAM 23G	⚠ 33 min 28 s VRAM 23G	⚠ 1.4 h VRAM 23G

Times are per whole clip (model load and text encoding included); the first run is slower due to kernel compilation. ⚠ = weight offloading / noticeably slower; ✕ = not recommended.

Reading the numbers

With the recommended combo, the green (no disk spill) ceiling is 416 × 256 · 10s at ~1 min 11 s per clip. To unlock 2.0MP and full durations: step up to a 32GB card (e.g. 5090), or use a smaller quant (GGUF Q3/Q4).

Recommended combo (Quality tier (24GB VRAM))

Main model: minimax_h3_fl2va_int8_convrot.safetensors（34.04GB）
Text encoder: qwen3vl_32b_minimax_h3_int8_convrot.safetensors（27.14GB）
Acceleration: LightX2V Turbo 8-step v1.0 · ~2.08× sampling
Launch flags: --fast-disk

768p-1080p 5-10s; 8-20 steps; 2-10 min per clip

24GB needs almost no layer-by-layer offloading; measured INT8 vs bf16 quality gap is about 0.9% quantization error

FAQ

Can the RTX 4090 24GB run MiniMax H3?

Yes. Estimated with 32GB RAM + fast-disk: 416×256·10s at ~1 min 11 s per clip. Configs that exceed VRAM fall back to weight offloading (slower, but they still render).

How fast is H3 on the RTX 4090 24GB?

About 1 min 11 s per 416×256·10s clip (including model load and encoding). Longer or denser clips scale quadratically in the attention term.

How much RAM do I need?

At least 64GB of system RAM recommended (minimum 32GB plus a 16GB pagefile). With 32GB, always enable --fast-disk or weights spill into the pagefile.

How can I make it faster?

Stack step distillation (e.g. LightX2V Turbo 8-step v1.0), enable SageAttention / Kitchen attention; for long clips try SLA / VSA sparse attention. All of it is switchable in step 2 of the calculator.

→ Run the numbers for your RAM and budget in the calculator (RTX 4090 24GB pre-filled)

Same tier: RTX 3090 24GB · RTX 4070 Ti Super 16GB · RTX 4080 16GB · RTX 4080 32GB (modded) · RTX 4080 Super 16GB · RTX 4080 Super 32GB (modded) · RTX 4090 D 24GB · RTX 4090 48GB (modded)