SELF-HOSTING # Qwen3.8-Flash-Next: Self-Hosting Estimate Qwen3.8-Flash-Next (Alibaba (Qwen)) 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**180B**Sets weight VRAM Active parameters**6B**Sets decode compute and speed License**other**[Official repo →](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) Public benchmarks**No public benchmark yet** Vendor API price**—**input / output per 1M tokens · [Token prices →](https://www.tandiu.cn/en/token-market) Vendor API output speed**—**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 **2priced as 1 × 8-GPU HGX node** Est. decode speed **1,333 tok/s**single-stream theoretical ceiling Est. batched throughput **out 2,267 tok/s**in 48,750 tok/s · many concurrent requests Hardware list price **$299,058–$498,429**Cloud rental (live) ×2: $15.77–$15.77/hr (2026-11-04) Hardware market price **$1,047,167** 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 **4** Est. decode speed **2,667 tok/s**single-stream theoretical ceiling Est. batched throughput **out 4,533 tok/s**in 108,333 tok/s · many concurrent requests Hardware list price **$119,772–$179,658** Hardware market price **$359,029–$718,058** 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 **FP16/BF16 (full precision)** GPUs needed **4priced as 1 × 8-GPU HGX node** Est. decode speed **2,667 tok/s**single-stream theoretical ceiling Est. batched throughput **out 4,533 tok/s**in 97,500 tok/s · many concurrent requests Hardware list price **$299,058–$498,429**Cloud rental (live) ×4: $29.79–$50.04/hr (2026-11-05) Hardware market price **$448,786–$523,584** 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 **FP16/BF16 (full precision)** GPUs needed **4priced as 1 × 8-GPU HGX node** Est. decode speed **1,600 tok/s**single-stream theoretical ceiling Est. batched throughput **out 2,720 tok/s**in 42,857 tok/s · many concurrent requests Hardware list price **$279,837–$499,708**Cloud rental (live) ×4: $11.96–$18.43/hr (2026-11-05) Hardware market price **$224,393–$233,369** 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 **FP16/BF16 (full precision)** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **2,233 tok/s**single-stream theoretical ceiling Est. batched throughput **out 3,797 tok/s**in 85,713 tok/s · many concurrent requests Hardware list price **$239,860–$831,514**Cloud rental (live) ×8: $15.47–$32.90/hr (2026-11-05) Hardware market price **$320,732** 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 **FP16/BF16 (full precision)** GPUs needed **8priced as 1 × 8-GPU HGX node** Est. decode speed **2,233 tok/s**single-stream theoretical ceiling Est. batched throughput **out 3,797 tok/s**in 85,713 tok/s · many concurrent requests Hardware list price **$293,207–$311,158** Hardware market price **$403,907–$418,867** 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 **FP16/BF16 (full precision)** GPUs needed **8** Est. decode speed **2,667 tok/s**single-stream theoretical ceiling Est. batched throughput **out 4,533 tok/s**in 12,827 tok/s · many concurrent requests Hardware list price **n/a** Hardware market price **$95,741–$131,644** 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 **FP16/BF16 (full precision)** GPUs needed **8** Est. decode speed **1,290 tok/s**single-stream theoretical ceiling Est. batched throughput **out 1,316 tok/s**in 16,224 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$119,676–$124,463** Hardware market price **$129,250–$143,612** 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 **FP16/BF16 (full precision)** GPUs needed **4** Est. decode speed **1,067 tok/s**single-stream theoretical ceiling Est. batched throughput **out 1,813 tok/s**in 32,587 tok/s · many concurrent requests Hardware list price **n/a** Hardware market price **$47,871–$119,676** 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 **FP16/BF16 (full precision)** GPUs needed **8** Est. decode speed **267 tok/s**single-stream theoretical ceiling Est. batched throughput **out 453 tok/s**in 27,733 tok/s · many concurrent requests Hardware list price **n/a** Hardware market price **$131,644–$143,612** 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 **FP16/BF16 (full precision)** GPUs needed **8** Est. decode speed **1,195 tok/s**single-stream theoretical ceiling Est. batched throughput **out 1,219 tok/s**in 26,000 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$128,200**Cloud rental (live) ×8: $9.62–$18.15/hr (2026-11-05) Hardware market price **$140,691** 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 **FP16/BF16 (full precision)** GPUs needed **8** Est. decode speed **932 tok/s**single-stream theoretical ceiling Est. batched throughput **out 951 tok/s**in 23,504 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$91,147** Hardware market price **$101,725–$107,709** 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 **INT8** GPUs needed **8** Est. decode speed **1,792 tok/s**single-stream theoretical ceiling Est. batched throughput **out 1,828 tok/s**in 13,416 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **n/a**Cloud rental (live) ×8: $4.49–$8.55/hr (2026-11-05) Hardware market price **$64,625** 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 **INT8** GPUs needed **8** Est. decode speed **1,152 tok/s**single-stream theoretical ceiling Est. batched throughput **out 1,175 tok/s**in 18,824 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$53,854–$77,790**Cloud rental (live) ×8: $4.81–$7.30/hr (2026-11-05) Hardware market price **$53,854–$77,790** Typical use SME teams building inference services, medium-low concurrency ## L20 48GB 48 GB · 864 GB/s · Compliant inference card Suggested precision **INT8** GPUs needed **8** Est. decode speed **1,152 tok/s**single-stream theoretical ceiling Est. batched throughput **out 1,175 tok/s**in 6,214 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$27,941–$28,101** Hardware market price **$41,887** 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 **INT8** GPUs needed **8** Est. decode speed **2,389 tok/s**single-stream theoretical ceiling Est. batched throughput **out 2,437 tok/s**in 21,788 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **n/a**Cloud rental (live) ×8: $3.54–$4.26/hr (2026-11-05) Hardware market price **$53,853** 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 **8** Est. decode speed **3,584 tok/s**single-stream theoretical ceiling Est. batched throughput **out 3,656 tok/s**in 15,444 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$19,745**Cloud rental (live) ×8: $3.54–$4.24/hr (2026-11-05) Hardware market price **$33,509–$47,871** 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 **8** Est. decode speed **2,688 tok/s**single-stream theoretical ceiling Est. batched throughput **out 2,742 tok/s**in 15,288 tok/s · many concurrent requests · PCIe multi-GPU penalty Hardware list price **$15,557–$22,737**Cloud rental (live) ×8: $2.70–$3.49/hr (2026-11-05) Hardware market price **$31,116–$35,903** 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.