stepfun
StepFun: Step 3.5 Flash
stepfun/step-3.5-flash
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Step 3.5 Flash is StepFun's open-weight text model. Its documentation describes a sparse Mixture-of-Experts design that activates about 11B of roughly 196B parameters per token, plus multi-token prediction for faster generation. The vendor card reports strong publisher-run results on reasoning, coding, and agentic benchmarks, a long context window, tool calling, and local deployment options; the OpenRouter listing confirms text-only I/O and mandatory reasoning. All performance figures are vendor-reported, so validate on your workload.[1][2]
Strengths
- Vendor-described sparse MoE design activates a small fraction of its total parameters per token, with multi-token prediction claimed to enable fast generation.[1][2]
- Publisher-run results on agentic and coding suites such as SWE-bench Verified and Terminal-Bench, plus a roughly 256K-token context window.[1][2]
- Apache-2.0 open weights with documented local deployment on vLLM, SGLang, Transformers, and llama.cpp.[2]
Limitations
- Every performance figure comes from the vendor's own model card; no independent evaluation was included in this source pack, so results need workload validation.[2]
- Text-only input and output, and the OpenRouter catalog marks reasoning as mandatory, which may add overhead for simple prompts.[1]
- Local inference demands substantial memory (the card lists roughly 120GB minimum for quantized weights), and the knowledge cutoff is not disclosed.[1][2]
Where to use it
- Agentic coding and tool-using assistants: worth piloting given tool-call support and vendor-reported software-engineering results.[1][2]
- Long-document or large-codebase analysis that can exploit the extended context window.[1][2]
- Self-hosted deployments where data must stay on-premises, per the documented local-inference paths.[2]
Sources
- [1] OpenRouter catalog · StepFun: Step 3.5 FlashSources checked: Nov 13, 2026
- [2] Model documentation · StepFun: Step 3.5 FlashSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....
Specifications
- Model ID
- stepfun/step-3.5-flash
- Creator
- stepfun
- Added to OpenRouter
- 29.01.2026
- Context length
- 262,144
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.100 / out $0.300
- Knowledge cutoff
- —
- Expiration date
- 15.11.2026
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: mandatory · parameters: include_reasoning, reasoning |
| Coding | No measurement | — |
| Agents | No measurement | parameters: tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: mandatory · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| SiliconFlowfp8 | 262,144 | $0.100 | $0.300 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.