Step-3.5-Flash
Official · BF16 · safetensors
Runtime certified: Transformers
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| BF16 | safetensors | 16 | Official | Runtime certified |
StepFun · Flash · Step 3.5
Step 3.5 Flash: Specs, GPU requirements, hosting & deployment
Step 3.5 Flash is a recent Flash release in the StepFun family, certified from its immutable official source revision.
Total parameters196.0B
Active parameters11.0B
Context window262,144 tokens
Release statusRecent · Recent
ArchitectureMoE Transformer
ModalitiesText
LicenseApache License 2.0
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed
Release date2026-02-01
Architecture, strengths and use cases
Step 3.5 Flash is an official self-hostable StepFun release in the StepFun Flash track. It uses MoE Transformer, carries 196B / 11B MoE, supports a verified context window of 262,144 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
Step 3.5 Flash uses MoE Transformer with 196B / 11B MoE. The stored architecture profile records 64 layers, hidden width 8192, 64 attention heads and a 262,144 token context ceiling where the official configuration publishes those fields.
Verified MoE Transformer architecture
262,144 token certified context
Official BF16 checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Recommended hardware
Use only the listed transformers runtime configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, runtime overhead, safety headroom, GPU architecture and topology before ranking rentable and reference hardware.
Estimated weight memory376.03 GB
KV cache2 GB
Runtime overhead95.8 GB
Total with safety margin587.55 GB
Technically compatible reference sizing
No current FusionPointAI inventory passes every runtime, architecture, topology and memory gate for this exact checkpoint. Use the estimator for global compatible hardware.
License and self-hosting considerations
The full 196B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
Step 3.5 Flash is published under Apache License 2.0. The release-scoped official license link and verification date are retained with the catalogue record.
Commercial Use Allowed. The official license permits commercial use; its obligations still apply.
Alternatives and sibling variants
Step 3.7 Flash, Step-3.7-Flash, Step-3.5-Flash
Release timeline
stepfun/stepfun/step-3-5-flash
Source details
Source revision: 861dcb086e44521b2c16b56c08fa7c05a3c19e06151bf96b75fba2ac6e1637c0
Frequently asked questions
How much VRAM does Step 3.5 Flash require?
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible runtime configurations.
Which runtimes support Step 3.5 Flash?
Certified upstream runtime evidence is available for transformers.
Can Step 3.5 Flash be used commercially?
The stored commercial classification is Commercial Use Allowed. Consult the official Apache License 2.0 terms for the final obligations.
Official references
Last verified: 2026-08-26