Step 5 Preview
Needs 781 GB at FP8 / INT8. The smallest real configuration today is 4× MI325X at Cyfuture AI, $12.29 per hour, about $8,972 a month at 730 hours.
- Per hour
- $12.29
- Per month
- $8,972
- Cheapest at
Cyfuture AI
What is documented about Step 5 Preview
news checked 2026-10-27StepFun's largest open release. Card: '600B total parameters, 27B active', up to 1M context, text and image input (the intro also mentions video). HF counts 604.3B. The model was announced around 18-20 Sept; the open weights were published 2026-10-15 (StepFun X post, MarkTechPost). The HF repo was created 2026-09-20. 352 experts, top-8; 72 sliding + 23 full-attention entries, including 3 MTP layers. Card serving guidance: FP8 runs on 4x B300 or 8x H200, BF16 on 8x B300 or 16x H200.
- https://huggingface.co/stepfun-ai/Step-5-Preview
- https://huggingface.co/stepfun-ai/Step-5-Preview/raw/main/config.json
- https://huggingface.co/api/models/stepfun-ai/Step-5-Preview?blobs=true
Memory · 2 precisions · 6 context lengthsHow much VRAM it needs
Weights plus runtime overhead, then the KV cache that grows with the context you serve.
| Precision | Weights | Weights + headroom | 4k context | 32k context | 128k context | 256k context | 1M context |
|---|---|---|---|---|---|---|---|
| FP8published checkpoint | 650 GB | 781 GB | 787 GB | 815 GB | 912 GB | 1,040 GB | 1,813 GB |
| INT4 / MXFP4estimated | 360 GB | 432 GB | 438 GB | 467 GB | 563 GB | 692 GB | 1,465 GB |
| KV cacheone sequence, FP16 cache, included above | — | — | 6 GB | 35 GB | 131 GB | 260 GB | 1,033 GB |
Smallest configuration per GPU · 110 GPUs · cheapest eligible on-demand offer per GPU count, every provider on fileWhat it costs to run, GPU by GPU
The fewest GPUs of one type whose total memory fits the model at short context, priced from the cheapest eligible on-demand offer for exactly that count (45-day window, newest reading per provider), with the next providers beside it. Monthly = 730 hours.
| GPU | Smallest config | VRAM used | $ / hour | $ / month | Cheapest at |
|---|---|---|---|---|---|
| MI325XBest fit | 4× 256 GB1024 GB total | 76% | $12.29 | $8,972 | Cyfuture AInot scored · read Nov 1 · receipt |
| MI300X | 8× 192 GB1536 GB total | 51% | $13.68 | $9,986 |  TensorWaveTrust 64 · read Nov 1 · receipt · also Cyfuture AI $22.38, AtmosCompute $26.77 |
| H200 | 8× 141 GB1128 GB total | 69% | $21.04 | $15,359 |  Airon AInot scored · read Nov 1 · receipt · also Geodd $21.76, Vast.ai $22.11 |
| B300 | 4× 288 GB1152 GB total | 68% | $23.29 | $17,004 |  UpCloudTrust 75 · read Nov 1 · receipt · also FPT AI Factory $27.96, Vast.ai $37.50 |
| MI355X | 8× 288 GB2304 GB total | 34% | $23.60 | $17,228 | TensorWaveTrust 64 · read Nov 1 · receipt |
| H200 NVL | 8× 141 GB1128 GB total | 69% | $26.09 | $19,042 |  IONOSnot scored · read Nov 1 · receipt · also Geodd $27.12, Vast.ai $39.47 |
| B200 | 8× 180 GB1440 GB total | 54% | $32.91 | $24,026 | UpCloudTrust 75 · read Nov 1 · receipt · also Airon AI $33.04, Deep Infra $39.12 |
| B300 CC | 4× 288 GB1152 GB total | 68% | $46.31 | $33,806 |  VerdaTrust 81 · read Nov 1 · receipt |
| GB300 | 4× 288 GB1152 GB total | 68% | $52.21 | $38,113 | VerdaTrust 81 · read Nov 1 · receipt |
| B200 CC | 8× 180 GB1440 GB total | 54% | $72.37 | $52,830 | VerdaTrust 81 · read Nov 1 · receipt |
Step 5 Preview · related