All models
Stepfun
Step 3.7 Flash fastruns locally
stepfun/step-3-7-flash
Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters...
Input / 1M
$0.20
USD per million tokens
Output / 1M
$1.15
USD per million tokens
Est. cost / token
$0.000000437
$0.44 per 1M, 3:1 mix
Context window
256K
max output 230K
Major features
VisionOpen weightsCachingReasoningJSON schemaTools
Task suitability (heuristic 1–5, not a benchmark)
- coding3
- reasoning3
- math3
- summarization4
- extraction4
- classification4
- chat3
- creative3
- translation4
- vision3
- long context3
- agentic3
Run it locally
Will it run on my machine? →| Quantization | Quality | Memory needed (8K ctx) |
|---|---|---|
| Q2_K | poor | 93.3 GB |
| Q3_K_M | fair | 108 GB |
| Q4_K_Mrecommended | good | 134 GB |
| Q5_K_M | very good | 155 GB |
| Q6_K | excellent | 178 GB |
| Q8_0 | near-lossless | 229 GB |
| F16 | lossless | 428 GB |
Estimates.
Price history
| Recorded | Input | Output | Cached | Source |
|---|---|---|---|---|
| 2026-11-05 03:20 | $0.20 | $1.15 | $0.040 | openrouter |
A row is added whenever a sync detects a price change. Vendor pricing source
Pricing (USD per 1M tokens)
- Input
- $0.20
- Output
- $1.15
- Cached input
- $0.040
- Blended (3:1)
- $0.44
Details
- Company
- Stepfun
- Version
- 3.7
- Family
- Step
- Released
- 2026-05-28
- Parameters
- 201.4B
- Max output
- 230K
- Open weights
- Yes
- Inputs
- image, text, video
- Last synced
- 46 min ago