Step-5-Preview
600B total parameters · 27B active parameters · Up to 1M context
Params
604.3 B
Context
—
Downloads 30d
2 K
Likes
127
Download history
tracking started — chart appears after 7 days of snapshots (5 recorded)2 K downloads in the last 30 days
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | f8_e4m3 | 650.3 GB | 806.5 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ curl -s https://aimodelscomparison.com/api/v1/models/step-5-preview
{
"hf_id": "stepfun-ai/Step-5-Preview",
"params_b": 604.34,
"context_length": null,
"license": { "id": "other", "commercial": "unknown" },
"downloads_30d": 2467,
"vram_estimates": [
{ "quant": "f8_e4m3", "gb": 806.5 }
],
"updated_at": "2026-10-19T06:02:11Z"
}
Use via API
via OpenRouter · USD per 1M tokens · synced 2026-10-24$1.00
input / 1M
$2.70
output / 1M
$0.10
cached input / 1M
$3.70
1M in + 1M out
- Context window
- 1,000,000 tokens
- Max output
- 64,000 tokens
- Providers
- 4
- Cheapest provider
- DeepInfra · $1.00 / $2.70
- Best throughput
- —
- AA Intelligence / Coding / Agentic
- 43.7 / — / —
✓ Tool calling
✓ Reasoning
✓ Structured outputs
✓ Image input
✗ File input
✗ Audio input
✓ Prompt caching
Specifications
- Architecture
- Step5ForConditionalGeneration
- Parameters
- 604.3 B
- Tensor type
- F8_E4M3
- Licence
- other
- First seen on the Hub
- 2026-09-20
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-20
Compare with any text-generation model