Skip to content
Model Pareto

Cosmos3-Super-Image2Video-4Step

NVIDIAOpen weightshuggingface.co ↗
Compare with nearest rivalsCompare with…
Price
—
Render
~91sper clipest.

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.

Hardware
1× B200 180GB
FP8 weights
Throughput
~2,601 tok/s
many requests batched
Price
$0.31–2.79 /M tok
busy → light use
On your own machine
Mac Studio M5 Max 128GB (int4, ~13 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (64B params) + 25% KV-cache headroom ≈ 80 GB
  • 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~2,601 output tok/s aggregate at batch 74 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~16,611 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Video generation

AA Text-to-Video ArenaEstimated
~#13 of 26 (estimated)~1224 Elo
  1. #1Wan 3.0Few votes1335
  2. #2Gemini Omni Flash1332
  3. #12Vidu Q3 Pro1225
  4. ~#13Cosmos3-Super-Image2Video-4StepEstimated1224
  5. #13Bach-1.0 Preview1223
No published score — estimated from related benchmarks
AA Text-to-Video Arena (with audio)Estimated
~#10 of 23 (estimated)~1096 Elo
  1. #1Gemini Omni Flash1233
  2. #2Wan 3.01230
  3. #9Wan 2.71098
  4. ~#10Cosmos3-Super-Image2Video-4StepEstimated1096
  5. #10SkyReels V41095
No published score — estimated from related benchmarks
LMArena Text-to-VideoEstimated
~#14 of 18 (estimated)~1334 Elo
  1. #1Gemini Omni 1.1 FlashFew votes1516
  2. #2Gemini Omni Flash1513
  3. #13Wan 2.71337
  4. ~#14Cosmos3-Super-Image2Video-4StepEstimated1334
  5. #14Seedance 1.5 pro1256
No published score — estimated from related benchmarks