SELF-HOSTING # Step 5 Preview: Self-Hosting Estimate Step 5 Preview (StepFun) on different GPUs: configuration, estimated performance and use cases. Data updated: 2026-09-16 [← All self-hosting estimates](https://www.tandiu.cn/en/local-deploy) Total parameters**604B**Sets weight VRAM Active parameters**27B**Sets decode compute and speed License**other**[Official repo →](https://huggingface.co/stepfun-ai/Step-5-Preview) Public benchmarks**AA 43.7 · LiveBench 62.0** Vendor API price**$1.05 / $2.99**input / output per 1M tokens · cached $0.105 · [Token prices →](https://www.tandiu.cn/en/token-market) Vendor API output speed**228.9 tok/s**first token 1.16s · measured by Artificial Analysis; compare with local estimates below ## B300 288GB 288 GB · 8,000 GB/s · Top-tier Blackwell card (no official channels to China) Suggested precision **FP16/BF16 (full precision)** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **1,185 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 2,015 tok/s**in 43,333 tok/s · many concurrent requests Hardware list price **$497,836**Cloud rental (live) ×8: $79.52–$120.01/hr (2026-10-24) Hardware market price **$1,048,061** Typical use Enterprise-grade ultra-large-scale training and high-concurrency inference clusters ## GB200 192GB (single card equivalent) 192 GB · 8,000 GB/s · Top-tier Blackwell card (no official channels to China) Suggested precision **FP16/BF16 (full precision)** GPUs needed **8** Est. decode speed **1,185 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 2,015 tok/s**in 48,148 tok/s · many concurrent requests Hardware list price **$239,748–$359,623** Hardware market price **$718,670–$1,437,341** Typical use Ultra-large-scale trillion-parameter model training and inference clusters (networked with Grace CPU) ## B200 180GB 180 GB · 8,000 GB/s · Top-tier Blackwell card (no official channels to China) Suggested precision **INT8** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **2,370 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 4,030 tok/s**in 43,333 tok/s · many concurrent requests Hardware list price **$240,769–$563,955**Cloud rental (live) ×8: $50.02–$60.02/hr (2026-10-24) Hardware market price **$449,169–$524,031** Typical use Enterprise-grade large-scale online inference and training ## H200 141GB 141 GB · 4,800 GB/s · Top-tier export-controlled card (channels to China are volatile) Suggested precision **INT8** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **1,422 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 2,418 tok/s**in 19,047 tok/s · many concurrent requests Hardware list price **$248,598–$451,084**Cloud rental (live) ×8: $27.41–$40.43/hr (2026-10-24) Hardware market price **$224,585–$233,568** Typical use Enterprise-grade production environment high-concurrency inference ## H100 SXM5 80GB 80 GB · 3,350 GB/s · Top-tier export-controlled card (no official channels to China) Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **1,985 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 3,375 tok/s**in 19,047 tok/s · many concurrent requests Hardware list price **$216,521–$834,005**Cloud rental (live) ×8: $17.09–$29.98/hr (2026-10-24) Hardware market price **$404,252–$419,224** Typical use Enterprise-grade production environment high-concurrency inference/training ## H800 SXM 80GB 80 GB · 3,350 GB/s · Top-tier export-controlled card (discontinued) Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **1,985 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 3,375 tok/s**in 19,047 tok/s · many concurrent requests Hardware list price **$293,457–$311,424** Hardware market price **$404,252–$419,224** Typical use Enterprise-grade production deployment (original China-compliant model, now discontinued) ## H20 96GB 96 GB · 4,000 GB/s · Export-compliant special supply card (significantly reduced compute power) Suggested precision **INT8** GPUs needed **8** Est. decode speed **1,185 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 2,015 tok/s**in 2,850 tok/s · many concurrent requests Hardware list price **n/a** Hardware market price **$95,823–$131,756** Typical use Suboptimal choice in export-compliant environments, medium-to-large scale inference deployment (FP16 compute power is only about 15% of H100, but higher memory bandwidth offers an advantage in decode-intensive scenarios) ## A800 PCIe 80GB 80 GB · 1,935 GB/s · Compliant special supply card (export compliance status adjusted multiple times) Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8** Est. decode speed **1,147 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 1,170 tok/s**in 3,605 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$119,778–$124,570** Hardware market price **$173,679** Typical use Medium-scale inference/training deployment in export-compliant environments ## Ascend 910C 128GB 128 GB · 3,200 GB/s · Huawei dual-chip data center card (domestically produced) Suggested precision **INT8** GPUs needed **8** Est. decode speed **948 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 1,612 tok/s**in 14,483 tok/s · many concurrent requests Hardware list price **n/a** Hardware market price **n/a** Typical use Domestically produced large model training and inference clusters, CloudMatrix and other super-node solutions ## Ascend 910B 64GB 64 GB · 400 GB/s · Domestic alternative (not subject to US export controls) Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8** Est. decode speed **237 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 403 tok/s**in 6,163 tok/s · many concurrent requests Hardware list price **n/a** Hardware market price **$131,756–$143,734** Typical use Domestic alternative, not subject to US export controls, medium-to-large scale inference/training deployment, memory bandwidth significantly lower than NVIDIA cards of similar VRAM capacity ## RTX PRO 6000 Blackwell 96GB 96 GB · 1,792 GB/s · Professional workstation/server card (China-specific version is 6000D 84GB) Suggested precision **INT8** GPUs needed **8** Est. decode speed **531 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 542 tok/s**in 5,778 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$128,309**Cloud rental (live) ×8: $9.62–$19.76/hr (2026-10-24) Hardware market price **n/a** Typical use Single-card large VRAM inference, local deployment of medium MoE models, private all-in-one machine ## RTX PRO 5500 Blackwell 84GB 84 GB · 1,398 GB/s · Professional workstation card (specs are NVIDIA preliminary data) Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8** Est. decode speed **828 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 845 tok/s**in 5,223 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **n/a** Hardware market price **$101,812–$113,789** Typical use Workstation large VRAM inference, departmental private deployment ## RTX PRO 5000 Blackwell 48GB 48 GB · 1,344 GB/s · Professional workstation card Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8** Est. decode speed **796 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 812 tok/s**in 2,981 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **n/a**Cloud rental (live) ×8: $4.85–$9.63/hr (2026-10-24) Hardware market price **$64,680** Typical use Small to medium model inference, development testing, single-machine multi-card private deployment ## L40S 48GB 48 GB · 864 GB/s · Compliant inference card Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8** Est. decode speed **512 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 522 tok/s**in 4,183 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$53,900–$77,856**Cloud rental (live) ×8: $3.33–$6.42/hr (2026-10-24) Hardware market price **$53,900–$77,856** Typical use SME teams building inference services, medium-low concurrency ## L20 48GB 48 GB · 864 GB/s · Compliant inference card Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **8** Est. decode speed **512 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 522 tok/s**in 1,381 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$27,965–$28,125** Hardware market price **$29,945–$41,922** Typical use SME teams building inference services, medium-low concurrency ## RTX 5090 32GB 32 GB · 1,792 GB/s · Consumer flagship (not sold in China, 5090D for domestic market) Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **16 (exceeds one node: multi-node needed)** Est. decode speed **2,124 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 2,166 tok/s**in 9,684 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **n/a**Cloud rental (live) ×16: $7.50–$10.74/hr (2026-10-24) Hardware market price **$83,845–$129,361** Typical use Local inference and development for individuals/small teams, multi-card VRAM pooling for quantized MoE ## RTX 5090D 24GB 24 GB · 1,344 GB/s · Consumer-grade/small-scale self-built Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **16 (exceeds one node: multi-node needed)** Est. decode speed **1,593 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 1,625 tok/s**in 6,864 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$39,524**Cloud rental (live) ×16: $7.95–$11.55/hr (2026-10-24) Hardware market price **$67,076–$95,823** Typical use Personal development/small-scale verification/POC, not recommended for production-level high concurrency ## RTX 4090D 24GB 24 GB · 1,008 GB/s · Consumer-grade/small-scale self-built Suggested precision **INT4 (AWQ/GPTQ)** GPUs needed **16 (exceeds one node: multi-node needed)** Est. decode speed **1,195 tok/s**single-stream theoretical ceiling · vendor API 229 tok/s Est. batched throughput **out 1,219 tok/s**in 6,795 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$31,140–$45,513**Cloud rental (live) ×16: $6.43–$7.78/hr (2026-10-24) Hardware market price **$62,285–$71,867** Typical use Personal development/small-scale verification/POC, not recommended for production-level high concurrency FAQ ## FAQ **How is the estimated decode speed calculated?** Estimated speed = (memory bandwidth × GPU count) ÷ (active parameters × bytes per parameter): the memory-bound upper limit for one request (roofline). It scales linearly with GPU count and ignores interconnect overhead, batching and framework losses, so real results are usually well below it. **Can I trust the speed numbers for consumer GPUs?** The linear assumption is reasonable for NVLink data-center cards (H/B series) but too optimistic for consumer PCIe cards without NVLink (RTX 5090D/4090D); multi-GPU consumer setups lose far more to communication, so equal-looking numbers do not mean equal performance. **How is “Suggested precision” chosen?** It is the highest precision that fits within 8 GPUs; beyond that it is flagged as multi-node. **Where do the prices come from?** USD converted at ¥6.68 per USD. Cloud rental is the international vast.ai market rate, not a China purchase price.