Step-3.5-Flash
Step-3.5-Flash is an open-weight language model from StepFun with 199B parameters. At 4-bit it needs about 135.1 GB of memory, so the smallest current device that fits it is the Mac Studio M5 Ultra (256GB).
Overview
Includes room for an 8k context and runtime overhead. More context needs more memory.
Key information
- Lab
- StepFun
- Released
- 1 Feb 2026
- Parameters
- 199B
- Type
- Language
- Context
- 262,144 tokens
- License
- Apache-2.0
- HF downloads, 30 days
- 176,088
- Repository
- stepfun-ai/Step-3.5-Flash
More from StepFun
View allWill it run on your hardware?
How we estimate| Estimate or device | Memory |
|---|---|
| Q4 | 135.1 GB |
| Q8 | 236 GB |
| BF16 | 442.2 GB |
| RTX 5060 Ti 16 GB | 16 GB usable |
| RTX 5090 32 GB | 32 GB usable |
| DGX Spark 128 GB | 108.8 GB usable |
| Mac Studio M5 Ultra 256 GB | 192 GB usable |
Single-device memory estimate. Pick a device to update the On your device column.
36 GB usable memory
| Format | Memory needed | On your device | Example hardware |
|---|---|---|---|
| 4-bit (Q4) | ≈ 135.1 GB | Too big 375% | Mac Studio M5 Ultra (256GB)256 GB memory |
| 8-bit (Q8) | ≈ 236 GB | Too big 656% | RubinAnnounced288 GB memory |
| 16-bit (BF16) | ≈ 442.2 GB | Too big 1228% | No single device in our directory has enough memory |
Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.
Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.
Where to run it
On a Mac
Start with the MLX files, built for Apple Silicon.
mlx-community/Step-3.5-Flash-4bitCheck your MacWith GGUF
Choose a file for llama.cpp or a compatible app.
ggml-org/Step-3.5-Flash-GGUFFind compatible hardwareThrough an API
SiliconFlow (fp8) is the cheapest provider we found.
$0.10 / $0.30
Input / output per 1M tokens
₹9.43 / ₹28 at ₹94.29 per $1
Blended: $0.15 (₹14) per 1M. Ranked by a 3:1 mix of input and output tokens.
mlx_lm.generate --model mlx-community/Step-3.5-Flash-4bit \
--prompt "Explain how a transformer works."llama-cli -hf ggml-org/Step-3.5-Flash-GGUF \
-c 8192 -p "Explain how a transformer works."Example commands built from the repository ids above. They need a working install of MLX or llama.cpp. We have not run them; check each repository for available files.
| Precision | |||||
|---|---|---|---|---|---|
| SiliconFlowCheapest | $0.15₹14 | $0.10₹9.43 | $0.30₹28 | 262,144 | FP8 |
INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.
More about this model
Frequently asked questions
How much memory does Step-3.5-Flash need? At an 8k context, about 135.1 GB at 4-bit, 236 GB at 8-bit and 442.2 GB at 16-bit (BF16). A longer context needs more.
Can I run Step-3.5-Flash on a Mac? The smallest current Mac that fits the 4-bit estimate of 135.1 GB is the Mac Studio M5 Ultra (256GB). An MLX build is available: mlx-community/Step-3.5-Flash-4bit.
Does the full context fit? Step-3.5-Flash supports up to 262,144 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.
What does it cost through an API? The cheapest provider we found is SiliconFlow, at $0.10 per 1M input tokens and $0.30 per 1M output tokens (₹9.43 and ₹28 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.
License in plain English
Apache-2.0. This license permits commercial use, modification and redistribution, with conditions such as keeping the required notices. This summary is not legal advice.
OMHQ RigBench
Not yet tested. We show no speed or quality score without a measurement.
Revision history and sources
Last updated . Sources:Hugging Face model card, MLX repository, GGUF repository, OpenRouter.