tokenstat
tokenstat

NVIDIA: Step 5 Preview

nvidia/step-5-preview · rates as of · How this catalog is built · Benchmarks

StepFun's next-generation flagship base model for coding and professional knowledge work, with native text, image, and video input and a 1M-token context window

Cheapest input
$1
$/MTok · models.dev
Cheapest output
$2.70
$/MTok
Context
1.0M
tokens
Sources
3
19 listings

Price changes

Cheapest list input/output only. Rates move rarely, so this is an event log. Down means cheaper. Showing the 8 latest changes.

Input $/MTok over sampled dates
Output $/MTok
  • ↓ lowerOutput $2.74 → $2.70
  • ↑ higherInput $0.959 → $1

Sources

Same model, different sources. Our first-party rate is listed first, then router and aggregator rows stay separate so you can see the spread instead of a single blended number. Free/$0 listings are hidden by default (3 omitted). Show free/$0 listings.

SourceProviderInputOutputCache readContext
models.devcheapeststep-5-previewEmpirioLabs AI$1$2.70$0.051.0M
models.devstep-5-previewDevPass (LLM Gateway)$1$2.70$0.051M
models.devstep-5-previewOpenCode Go$1$2.70$0.11M
models.devstep-5-previewRequesty$1$2.70$0.051.0M
models.devstepfun/Step-5-PreviewCommandCode$1$2.70$0.051M
models.devstepfun/step-5-previewKilo Gateway$1$2.70$0.11M
models.devstepfun/step-5-previewNanoGPT$1$2.70$0.051M
models.devstepfun/step-5-previewNovita AI$1$2.70$0.11M
OpenRouterstepfun/step-5-previewOpenRouter$1$2.70$0.11.0M
models.devstepfun/step-5-previewVercel AI Gateway$1$2.70$0.11.0M
models.devstepfun-ai/Step-5-PreviewDeepInfra$1$2.70$0.11.0M
models.devdeepinfra/stepfun-ai/Step-5-PreviewEden AI$1$2.70$0.11.0M
models.devtencent/step-5-previewLLM Gateway$1$2.70$0.051M
models.devstep-5-previewStepFun (Global)$1$2.70$0.11.0M
LiteLLMstep-5-previewN/A$1$2.70$0.1N/A
models.devstep-5-previewStepFun (China)$1.05$2.99$0.10461M
models.devstepfun/step-5-previewAI/ML API$1.06$2.86$0.05291M
models.devstep-5-previewAIHubMix$1.08$3.09$0.0541M
LiteLLMkyma/step-5-previewN/A$1.35$3.65$0.135N/A

Capabilities

  • Reasoning
  • Tool calling
  • Structured output
  • Attachments
  • Open weights
  • In: text
  • In: image
  • In: video
  • In: audio
  • Out: text
Released: 2026-09-16Output limit: 64K tokens

Scores

Sparse coverage on purpose. Missing a score means we have not linked one yet, not that the model is unranked. Hover or tap the i next to any score for its source. Full boards on /benchmarks.

Arena rows are community-vote Elo from Arena AI, Artificial Analysis indices arrive via OpenRouter, and launch-eval figures are collected from vendor announcements, analyst writeups, and community benchmarks. Full boards and credits on /benchmarks.

See what this model costs you

The app reads the sessions your tools already wrote and prices this model's tokens at the published API rate. Free on your computer.