Get started

Step 5 Preview

Timed at 29.8 tok/s and scored 28.5 on the Intelligence Index.

retrieved sha256 · 0bb53a77713f

Models/StepFun/Step 5 PreviewUSD per 1M tokens · as OpenRouter routes it
stepfun/step-5-previewStepFun · base listing · Other tokenizer
OpenRouter list price · the routed rate, as the catalog carried itretrieved 17 Nov 16:29 UTC from openrouter.ai ↗
Input$1.00per 1M tokens
Output$2.70per 1M tokens
Cached input$0.10per 1M tokens
Context1M1,000,000 tokens
What the list says · the fields OpenRouter publishes for this id10 fields
AuthorStepFun
Listed
Canonical idstepfun/step-5-preview-20261008
Context1,000,000 tokens
Max output64,000 tokens
Modalitiestext, image, video → text
TokenizerOther
Reasoningalways on · effort high, medium, low, medium by default
Moderationnot moderated by OpenRouter
What the weights say · read from the model's own repository on Hugging Faceretrieved from huggingface.co ↗
Licenceother ↗
Parameters604B · 604,339,504,832 declared
Model classStep5ForConditionalGeneration
Downloadnot gated
Published

Measured speed

676 requests · 2 endpoints · 30-minute window
Throughput29.8 tok/sMedian output rate, weighted across the 2 endpoints serving it by the requests each answered.the speed board
First token2.95sMedian wait before the first token, over the same window and weighted the same way.how Vitna reads a figure
Endpoint spread24 to 85The slowest and the fastest route serving this model, in tokens per second. Speed is a property of the route, not of the weights.the board, fastest first

OpenRouter's own timing of its routed traffic, weighted by requests, over one 30-minute window read . Every route is on the endpoint ledger below.

Every endpoint

2 priced · 2 timed · cheapest first
  1. Featherlessfeatherless/fp4 · listed$1.00$2.70$0.10fp433k24 tok/s212.77s612 req98.50%
  2. StepFunstepfun/fp8 · listed$1.00$2.70$0.10fp864k85 tok/s234.65s64 req97.20%

Each track runs from a route's p90, the rate nine in ten of its requests met or beat, up to its median. Weighted by requests, the figure above reports where the traffic went rather than what one route does: 1 of the 2 timed routes are slower than it, and the middle route runs at 54.5 tok/s. The widest median-to-p90 gap is StepFun's, 3.7x.

Uptime is OpenRouter's own figure for each seller over the trailing day, read once a day, and the column prints the latest reading, taken . 3 of the 6 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: GMICloud at 0.00% on 3 Nov 2026, Featherless at 1.13% on 13 Nov 2026, and StepFun at 97.20% on 21 Nov 2026. Source: OpenRouter's public endpoints API, one reading a day for each seller: 1,211 sellers carried one on the latest of the 23 days held.

Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. List prices, not what anybody paid; a discount is the seller's own. Read . Speed is OpenRouter's own timing over one 30-minute window, read .

What each seller accepts

19 parameters · 2 sellers
  1. frequency_penaltyevery seller
  2. include_reasoningevery seller
  3. max_tokensevery seller
  4. reasoningevery seller
  5. reasoning_effortevery seller
  6. response_formatevery seller
  7. stopevery seller
  8. structured_outputsevery seller
  9. temperatureevery seller
  10. toolsevery seller
  11. top_pevery seller
  12. logprobs1 of 2not Featherless
  13. min_p1 of 2not StepFun
  14. presence_penalty1 of 2not StepFun
  15. repetition_penalty1 of 2not StepFun
  16. seed1 of 2not StepFun
  17. tool_choice1 of 2not StepFun
  18. top_k1 of 2not StepFun
  19. top_logprobs1 of 2not Featherless

11 of these 19 are accepted by every one of the 2 sellers that publish a list; the other 8 are declared by some sellers and not others, so what a request sending one of them gets depends on which seller serves it. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.

Price position

342 priced models

Blended $1.425 per million tokens, 3 in to 1 out: cheaper than 124 of the 342 priced models in the catalog. Each bar is one stretch of the catalog on a logarithmic price scale; the mark is this listing.

Measured quality

145 models with an index · 132 on a design arena board
Intelligence Index28.5Position 29 of 58, above 50% of the models scored.the intelligence leaderboard
Coding Indexnot scoredArtificial Analysis has not scored this model on its Coding Index.the coding leaderboard
Agentic Indexnot scoredArtificial Analysis has not scored this model on its Agentic Index.the agentic leaderboard

Not ranked in any Design Arena category.

Epoch Capabilities Index · a fit by a third party over every benchmark this model has results onthe whole index, with intervals

Epoch AI has not placed Step 5 Preview on its Capabilities Index, or places it under a name the pairing does not reach. Epoch places a model once four of its benchmarks carry a result.

Benchmark results

none of 6 boards

None of the 6 openly licensed boards Vitna mirrors has published a result for Step 5 Preview under a name that matches its listing, and names are matched to the catalog exactly and never guessed. Every board, with its licence and what its publisher measures, is on /benchmarks.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Source: each model's own Hugging Face repository (huggingface.co), as of 20 Nov 2026. 162 repositories read, 4 did not answer.
Licence, parameter count, model class and gating: each model's own Hugging Face repository, the one OpenRouter names for the listing.
A licence is the tag the uploader put on the repository, read from nothing else and verified by nobody.
A parameter count is the total the repository's safetensors files declare; for a mixture of experts that is every expert, not the share one token runs through.
Vitna ran none of these evaluations and adjusted none of these figures.