InferIndex
Language, English, EN
Account

ModelsStepFunStep 5 Preview

Step 5 Preview

stepfun/step-5-preview262k context Compare →
MedianThe median blended price among comparable offers: one price per provider (its cheapest route), standard tier, current prices only. Priority tiers and prices converted from points are not counted.
$1.43
Most expensiveThe highest blended price among comparable offers: one price per provider (its cheapest route), standard tier, current prices only. Priority tiers and prices converted from points are not counted.
$1.54
Offers
22
Providers
13

Cheapest right now

Featherless

directYou buy access straight from the company that runs the model on its servers. · fp4 · 262k

$1.43 blended / 1M tokens

No change7D
in
$1.00
out
$2.70
Cached input
–

Public leaderboards

  • Text Arena, style controlThe series the source publishes by default for text since 16 May 2025: ranks corrected for the style of the answers (style control). #63 of 419Best rank in the source's rank spread, 63rd to 102nd, given the confidence intervals of all ranked models. 82nd by score. 1,448 1,442 to 1,455Arena Score · 11,355 votes
  • Code Arena (WebDev)The Code Arena ranking (WebDev). The source publishes no style-corrected series (as read on 9 October 2026).High setting #36 of 152Best rank in the source's rank spread, 36th to 43rd, given the confidence intervals of all ranked models. 42nd by score. 1,571 1,562 to 1,580Arena Score · 5,966 votes listed as “Step 5 Preview (High)”
  • Agent Arena aggregateThe Agent Arena aggregate score: a score of another kind (not an Arena score), on a scale of its own; it cannot be compared with the other series. #32 of 57Best rank in the source's rank spread, 32nd to 43rd, given the confidence intervals of all ranked models. 37th by score. −0.027 −0.040 to −0.013IPS score · 2,555,544 observations

Source: Arena (opens in a new tab) · Data under the CC BY 4.0 license (creativecommons.org/licenses/by/4.0) · CC BY 4.0 (opens in a new tab) · scores and ranks as published on 28 Oct 2026; the presentation is ours

Step 5 Preview costs from $1.43 per million tokens (at a 3:1 input:output mix): the lowest current price, at 7 providers, checked on ; 22 offers are listed.

Listed offers, by price

Comparable offersOther offersSolid steps are comparable offers; dashed steps are the other offers, which are not counted in cheapest, median or most expensive.

Offers not counted in cheapest, median or most expensive (14)

  • $1.43 · DeepInfra via Eden AI · cannot be designated the cheapest
  • $1.43 · DeepInfra via Vercel AI Gateway · cannot be designated the cheapest
  • $1.43 · Featherless via Kilo Gateway · cannot be designated the cheapest
  • $1.43 · GMICloud via Kilo Gateway · cannot be designated the cheapest
  • $1.43 · StepFun via Kilo Gateway · cannot be designated the cheapest
  • $1.43 · Tencent via Eden AI · cannot be designated the cheapest
  • $1.43 · Glama via Glama · cannot be designated the cheapest
  • $1.43 · Nous Portal via Nous Portal · cannot be designated the cheapest
  • $1.43 · Tencent via LLM Gateway · another route of the same provider
  • $1.43 · DeepInfra via Requesty · cannot be designated the cheapest
  • $1.43 · Featherless via OpenRouter · another route of the same provider
  • $1.43 · GMICloud via OpenRouter · cannot be designated the cheapest
  • $1.43 · StepFun via OpenRouter · another route of the same provider
  • $1.54 · ShengSuanYun via ShengSuanYun · cannot be designated the cheapest

Cheapest that confirms no training on your prompts. The provider's published policy: prompts are not used for training.

$1.43 blended / 1M tokens

in
$1.00
out
$2.70
Cached input
$0.05

Price historyThe cheapest offer each day, against the lab's official list price when it publishes one. Each day at 00:00 UTC, the offer the API would have selected as the cheapest, with the same exclusions as /cheapest. Blended price, promotions included.

Cheapest offer and official list price, by day, 30 days
DayCheapest offerOfficial list price
2026-10-02$1.43, EmpirioLabs AI–
2026-10-03$1.43, EmpirioLabs AI–
2026-10-04$1.43, EmpirioLabs AI–
2026-10-05$1.43, EmpirioLabs AI–
2026-10-06$1.43, Tencent–
2026-10-07$1.43, Tencent–
2026-10-08$1.43, Tencent–
2026-10-09$1.43, StepFun–
2026-10-10$1.43, StepFun–
2026-10-11$1.43, StepFun–
2026-10-12$1.43, StepFun–
2026-10-13$1.43, StepFun–
2026-10-14$1.43, Novita–
2026-10-15$1.43, Novita–
2026-10-16$1.43, DeepInfra–
2026-10-17$1.35, DeepInfra–
2026-10-18$1.43, DeepInfra–
2026-10-19$1.43, Featherless–
2026-10-20$1.43, Featherless–
2026-10-21$1.43, Featherless–
2026-10-22$1.43, Featherless–
2026-10-23$1.43, Featherless–
2026-10-24$1.43, DeepInfra–
2026-10-25$1.43, DeepInfra–
2026-10-26$1.43, DeepInfra–
2026-10-27$1.43, DeepInfra–
2026-10-28$1.43, DeepInfra–
2026-10-29$1.43, DeepInfra–
2026-10-30$1.43, Featherless–
2026-10-31$1.43, Featherless–

Cheapest offer

Offers

22 of 22 offers

Offers for Step 5 Preview, one row per offer, sorted by blended price.
RankProvider · viaYou buy access through an aggregator: one account and one API that forward your requests to other providers. "Via" names the aggregator; the price is the one it charges you. When "via" says "hosted API", no aggregator is involved: the model belongs to a lab and another provider sells it.Input $/1MOutput $/1MBlended $/1MA single price that mixes input and output: 3 input tokens for 1 output token, per million tokens. It makes offers easy to rank; your real cost depends on your own mix.PrecisionHow finely the model's numbers are stored on the provider's servers. Lower precision (fp8, fp4) is cheaper to run and can slightly change answer quality; "unknown" means the provider doesn't say. ContextThe maximum amount of text, counted in tokens, the model can take in and write in a single request. Past this limit, the request is refused or cut.ConditionsWhat a provider says about how it handles your data and who can sign up: training on prompts, processing region, sign-up requirements. Each comes from the provider's published documents; when nothing is published, it stays unknown.SignalsOffer
1 Featherless 1.00 2.70 1.43 fp4 262k
2 EmpirioLabs AI 1.00 2.70 1.43 – 1M
3 NanoGPT 1.00 2.70 1.43 – 1M
4 Novita 1.00 2.70 1.43 – 1M
5 DeepInfravia Eden AI 1.00 2.70 1.43 fp8 1M United States Unverified. No public page for this offer: listed, not designated cheapestRetired: This seller announced it would stop serving the model · Retired Oct 29, 2026 · announced on deepinfra.com · read on Oct 29, 2026 (opens in a new tab)
6 DeepInfravia Vercel AI Gateway 1.00 2.70 1.43 fp8 1M United States Retired: This seller announced it would stop serving the model · Retired Oct 29, 2026 · announced on deepinfra.com · read on Oct 29, 2026 (opens in a new tab)
7 Featherlessvia Kilo Gateway 1.00 2.70 1.43 fp4 262k Unverified. No public page for this offer: listed, not designated cheapest
8 GMICloudvia Kilo Gateway 1.00 2.70 1.43 fp8 1M Unverified. No public page for this offer: listed, not designated cheapest
9 Stepfunvia Vercel AI Gateway 1.00 2.70 1.43 fp8 1M
10 StepFunvia Kilo Gateway 1.00 2.70 1.43 fp8 1M Unverified. No public page for this offer: listed, not designated cheapest
11 Tencentvia Eden AI 1.00 2.70 1.43 – 1M Unverified. No public page for this offer: listed, not designated cheapest
12 Tencentvia Vercel AI Gateway 1.00 2.70 1.43 – 1M
13 Glama 1.00 2.70 1.43 – 1M Unverified. No public page for this offer: listed, not designated cheapest
14 Nous Portal 1.00 2.70 1.43 – 1M Unverified. Neither seller nor price found on its page for 72 hours: listed, not designated cheapest
15 Tencentvia LLM Gateway 1.00 2.70 1.43 – – Fee 5%: LLM Gateway adds a 5% fee on credit usage, plus 1.5% for cards issued outside the US (read on llmgateway.io/pricing on Oct 30). Offers are ranked with the 5% included.
16 DeepInfravia Requesty 1.00 2.70 1.43 fp8 1M United States Retired: This seller announced it would stop serving the model · Retired Oct 29, 2026 · announced on deepinfra.com · read on Oct 29, 2026 (opens in a new tab)Fee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.
17 Featherlessvia OpenRouter 1.00 2.70 1.43 fp4 262k Fee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.
18 GMICloudvia OpenRouter 1.00 2.70 1.43 fp8 1M Fee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.
19 StepFunvia OpenRouter 1.00 2.70 1.43 fp8 1M Fee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.
20 ZenMux 1.05 3.00 1.54 – 1M
21 PPIO 1.05 3.01 1.54 – 1M
22 ShengSuanYun 1.05 3.01 1.54 – 256k Unverified. No public page for this offer: listed, not designated cheapest

No training on promptsPrompts may be used for trainingTraining terms not published

backend not named. The router bills this price itself without naming who serves the model.

Blended = (3 × input + output) / 4, per million tokens. Prices in USD, converted at the ECB rate when the provider bills in another currency.

Method

Monthly cost for your workload

Computed by the API for the three cheapest offers, per 30-day month.
  1. Featherless$100.50per month
  2. EmpirioLabs AI$100.50per month+$0.00 vs the cheapest
  3. NanoGPT$100.50per month+$0.00 vs the cheapest

Host it yourself instead? See the break-even

Compare Step 5 Preview with

Related models

More from StepFun

All StepFun models

Same index family

Use this data

The index is queryable. The API is free to use.
Free API
curl https://api.inferindex.dev/cheapest?model=stepfun/step-5-preview
MCP server
claude mcp add --transport http inferindex https://mcp.inferindex.dev/mcp

Also works in Cursor and Codex: see the MCP page.