Step-3.5-Flash-FP8
Support summary
Step-3.5-Flash-FP8 is a checkpoint published by StepFun, built on the Step3p5ForCausalLM architecture. Official recipes or cookbooks cover vLLM, without pinning a stable release.
- Documented quantization
- FP8
- Configured context
- 262,144 tokens
- License
- apache-2.0
Hardware, quantization and features count only statements about this checkpoint; engine-wide capabilities are listed with each release.
Hugging Face model card
- Publisher
- StepFun
- Task
- text generation
- License
- apache-2.0
- Library
- transformers
- Architecture
- Step3p5ForCausalLM
- Weight tensor elements
- 199.4B
- Tensor types
- F32, BF16, F8_E4M3
- Quantization
- fp8
- Configured context
- 262,144 tokens
- Base models
- stepfun-ai/step-3.5-flash
- Weight access
- Public
- Repository created
- 2026-02-01
Engine support
3 engines
| Engine | Release | Checkpoint support | Evidence scope | Last known supported |
|---|---|---|---|---|
| vLLM | Version not pinnedRecipe / cookbook | Documented recipeSource: vllm · models/stepfun-ai/Step-3.5-Flash.yamlReviewed | No supported release indexed | |
Features, dependencies and evidence for vLLM 0.28.0 | ||||
| SGLang | 0.5.19Newest indexed | No matching evidence curated | No supported release indexed | |
Features, dependencies and evidence for SGLang 0.5.19 | ||||
| TensorRT-LLM | 1.2.1Newest indexed | No matching evidence curated | No supported release indexed | |
Features, dependencies and evidence for TensorRT-LLM 1.2.1 | ||||