TEXT REASONING256K CONTEXT196B TOTAL11B ACTIVETOOLS
Step 3.5 Flash API Pricing
StepFun lists Step 3.5 Flash as a 196B-total / 11B-active sparse MoE language-reasoning model with 256K context. StepFun's USD price list charges $0.10/M input, $0.30/M output, and $0.02/M cached input.
Input - per 1M tokens
$0.10/M
Source StepFun USD list cache miss
Output - per 1M tokens
$0.30/M
Context 256K
Cached input - per 1M tokens
$0.02/M
Prompt cache hit row 80% off
Effective - agentic blend
$0.0556/M
92/8 split - 82% cache
§ 01 / TERMINAL
Run the numbers.
Calculator pre-loaded with the rates on StepFun's USD price list for Step 3.5 Flash, the text-reasoning Flash tier.
$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS
Real-world presets.
CODING AGENT
Repo repair task
$0.00266/task
OFFICE
Analyst memo draft
$0.00228/memo
SEARCH
Web research brief
$0.00176/brief
CHATBOT
Product support turn
$0.00051/turn
Estimate · stepfun-tokenizer-estimate · ≈3.85 chars/token Auto-counts as you type
This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.
Characters 264
Words 46
Tokens (estimated) 69 tokens
Cost as input · uncached $0.00001 USD
Cost as output · uncached $0.00002 USD
Cost as cached input $0.000001 USD
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Step 3.7 Flash | $0.20 cache $0.04 | $1.15 | $0.155 +2.8× | 256K | Cheaper StepFun multimodal agents |
| Step 5 Preview | $1.00 cache $0.05 | $2.70 | $0.419 +7.5× | 1M | StepFun's 1M-context agent flagship |
| Step 3.5 Flash Current | $0.10 cache $0.02 | $0.30 | $0.0556 agentic 92/8 | 256K | Cheap StepFun text reasoning |
| MiMo-V2.6-Flash | $0.14 cache $0.0028 | $0.28 | $0.0477 −14% | 1M | Very low-cost 1M-context MiMo row |
| MiMo-V2.6-Pro | $0.435 cache $0.0036 | $0.87 | $0.144 +2.6× | 1M | MiMo Pro reasoning row |
| GLM-4.7 | $0.60 cache $0.11 | $2.20 | $0.358 +6.4× | 200K | Coding agent backend |
| DeepSeek V4 Pro | $1.32 cache $0.044 | $3.96 | $0.569 +10× | 1M | Low-cost reasoning workloads |
Frequently asked.
Practical questions about Step 3.5 Flash pricing, cache hits, context size, and low-cost StepFun reasoning workloads.
Q · 01 What is Step 3.5 Flash priced at? +
StepFun's USD price list (platform.stepfun.ai) charges
$0.10/M input on a cache miss, $0.02/M on a cache hit, and $0.30/M output for step-3.5-flash. The Chinese platform lists a separate CNY 0.7 / 0.14 / 2.1 card; the two are set independently, so this page reads the dollar list rather than converting the yuan one.Q · 02 How is the effective price calculated? +
The headline effective tile uses the site standard agentic blend:
92% input, 8% output, and 82% input cache hits. For Step 3.5 Flash, that lands at about $0.06/M blended tokens.Q · 03 What context window does Step 3.5 Flash support? +
StepFun's model docs list
256K context for step-3-5-flash. The model page also lists a sparse MoE architecture with 196B total / 11B active.Q · 04 Does this page model media/audio/image pricing? +
No. StepFun also has vision, image, and audio model rows. This page covers the text-token pricing for the listed reasoning model, not image generation or speech pricing.
Q · 05 When was this price last checked? +
This page was verified against StepFun's USD price list on
2026-09-29. Until then it converted the CNY card, which overstated the rates; see the price history and the corrections log.