StepFunVideo model
Step-Video-TI2V 30B
Clip lengthn/a
n/alongest single cliphigher is better
Summary
Step-Video-TI2V 30B is a video model from StepFun. Step-Video-TI2V 30B handles text and images in one video model; the model description lists up to 102 frames.
Head-to-head comparisons
vs.Step-Video-TI2V 30B vs. FLUX.3 VideoNot enough data for a result.Step-Video-TI2V 30BFLUX.3 VideoImage to videon/a1,449Text to videon/a1,4935 s clipn/a$0.85vs.Step-Video-TI2V 30B vs. Dreamina Seedance 2.5 720pNot enough data for a result.Step-Video-TI2V 30BDreamina Seedance 2.5 720pImage to videon/a1,477Text to videon/a1,4745 s clipn/an/avs.Step-Video-TI2V 30B vs. Dreamina Seedance 2.0 720pNot enough data for a result.Step-Video-TI2V 30BDreamina Seedance 2.0 720pImage to videon/a1,475Text to videon/a1,4795 s clipn/a$0.76vs.Step-Video-TI2V 30B vs. MiniMax H3Not enough data for a result.Step-Video-TI2V 30BMiniMax H3Image to videon/a1,495Text to videon/a1,4605 s clipn/a$0.65vs.Step-Video-TI2V 30B vs. Veo 3.1Not enough data for a result.Step-Video-TI2V 30BVeo 3.1Image to videon/a1,398Text to videon/a1,3645 s clipn/a$2.00vs.Step-Video-TI2V 30B vs. Veo 3.1 FastNot enough data for a result.Step-Video-TI2V 30BVeo 3.1 FastImage to videon/a1,385Text to videon/a1,3625 s clipn/a$0.50
Show all comparisons (94)
All specifications
Architecture
- Checkpoint
- Distinct public weights in the official `stepfun-ai/Step-Video-TI2V` repository.
- Source: Checkpoint
Licence and access
Further details
- Input modes
- Text- and image-to-video in one 30B model; the official model description states up to 102 frames.
- Source: Input modes
- Undocumented maxima
- The model description documents 102 frames but no maximum output resolution; maxDurationSeconds remains null and frames are not converted to seconds.
- Source: Undocumented maxima
Sources and data date
| Type | Evidence and data date |
|---|---|
| Additional source | StepFun (retrieved November 1, 2026) · StepFun · Editorial description reviewed November 1, 2026 |