Home / GPU guide / RTX 5090 D V2 24GB

RTX 5090 D V2 24GB for MiniMax H3: Speed, VRAM & Recommended Setup

24GB GDDR7 · bandwidth 1344 GB/s · Blackwell · sm_120 · estimated with 32GB RAM / fast-disk · updated 2026-09-18

Verdict: Recommended: 416 × 256 · 10s, ~1 min 17 s per clip (VRAM peak ~23GB). See the table for higher tiers.
Size (16:9)5s8s10s15s
416 × 256
0.1MP · flicker-prone
1 min 12 s
VRAM 23G
1 min 15 s
VRAM 23G
1 min 17 s
VRAM 23G
⚠ 1 min 30 s
VRAM 23G
864 × 480
0.4MP
⚠ 1 min 44 s
VRAM 23G
⚠ 2 min 26 s
VRAM 23G
⚠ 3 min 3 s
VRAM 23G
⚠ 5 min 2 s
VRAM 23G
960 × 544
0.5MP
⚠ 2 min 1 s
VRAM 23G
⚠ 3 min 4 s
VRAM 23G
⚠ 3 min 60 s
VRAM 23G
⚠ 7 min 5 s
VRAM 23G
1152 × 640
0.7MP
⚠ 2 min 42 s
VRAM 23G
⚠ 4 min 42 s
VRAM 23G
⚠ 6 min 28 s
VRAM 23G
⚠ 12 min 45 s
VRAM 23G
1376 × 768
1.0MP · above native
⚠ 4 min 6 s
VRAM 23G
⚠ 8 min 4 s
VRAM 23G
⚠ 11 min 47 s
VRAM 23G
⚠ 25 min 47 s
VRAM 23G
1920 × 1088
2.0MP · above native
⚠ 11 min 48 s
VRAM 23G
⚠ 29 min 13 s
VRAM 23G
⚠ 47 min 18 s
VRAM 23G
✕ 2.0 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.

Recommended combo (Quality tier (24GB VRAM))

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 5090 D V2 24GB run MiniMax H3?

Yes. Estimated with 32GB RAM + fast-disk: 416×256·10s at ~1 min 17 s per clip. Configs that exceed VRAM fall back to weight offloading (slower, but they still render).

How fast is H3 on the RTX 5090 D V2 24GB?

About 1 min 17 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 5090 D V2 24GB pre-filled)

Same tier: RTX 5090 32GB · RTX 5090 D 32GB

Related: Hardware guide · Acceleration compared · 8GB VRAM setups