Qwen · Code · 480B parameters · 35B active · apache-2.0
Qwen3 Coder 480B A35B Instruct
Needs 576 GB at FP8 / INT8. The smallest real configuration today is 8× RTX PRO 6000 at UpCloud, $7.78 per hour, about $5,676 a month at 730 hours.
Parameters480B35B active per token
Context256k tokensmaximum published
Checkpoint960 GBBF16, as published
Licenseapache-2.0check before commercial use
Released—documented date
Smallest configuration · FP8 / INT8
8× RTX PRO 6000768 GB VRAM · 75% used
- Per hour
- $7.78
- Per month
- $5,676
- Cheapest at
UpCloud
Memory · 3 precisions · 5 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 |
|---|---|---|---|---|---|---|
| BF16published checkpoint | 960 GB | 1,153 GB | 1,154 GB | 1,161 GB | 1,186 GB | 1,219 GB |
| FP8 / INT8estimated | 480 GB | 576 GB | 578 GB | 585 GB | 610 GB | 643 GB |
| INT4 / MXFP4estimated | 288 GB | 346 GB | 347 GB | 354 GB | 379 GB | 413 GB |
| KV cacheone sequence, FP16 cache, included above | — | — | 2 GB | 9 GB | 34 GB | 67 GB |
Cross-checked on Hugging Face Oct 28, 2026: 480.2B parameters in the safetensors headers, 960.3 GB of weight files, attention layout from config.json (62 layers, 8 KV heads × 128). Method: weights as published at the native precision, verified on Hugging Face 2026-10-28; other precisions estimated at 1 (8-bit) and 0.6 (4-bit) bytes per parameter; × 1.2 runtime headroom; KV cache = 4096 bytes per token per layer × 62 full-attention layers (FP16 cache, one sequence).config.json ↗
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 |
|---|---|---|---|---|---|
| RTX PRO 6000Best fit | 8× 96 GB768 GB total | 75% | $7.78 | $5,676 | UpCloudTrust 75 · read Nov 1 · receipt · also Vast.ai $10.14, Lium $12.80 |
| Gaudi 2 | 8× 96 GB768 GB total | 75% | $8.27 | $6,036 |  Cyfuture AInot scored · read Nov 1 · receipt |
| A100 80GB | 8× 80 GB640 GB total | 90% | $9.82 | $7,166 | Edgevananot scored · read Nov 1 · receipt · also GPU.ai $10.00, io.net $10.50 |
| MI300X | 4× 192 GB768 GB total | 75% | $11.31 | $8,257 | Cyfuture AInot scored · read Nov 1 · receipt · also Hot Aisle $11.96, AtmosCompute $13.38 |
| B300 | 2× 288 GB576 GB total | 100% tight | $11.65 | $8,506 | UpCloudTrust 75 · read Nov 1 · receipt · also FPT AI Factory $13.98, Vast.ai $17.00 |
| MI325X | 4× 256 GB1024 GB total | 56% | $12.29 | $8,972 | Cyfuture AInot scored · read Nov 1 · receipt |
| H100 | 8× 80 GB640 GB total | 90% | $13.67 | $9,982 |  GcoreTrust 80 · read Nov 1 · receipt · also Lium $14.36, Airon AI $15.04 |
| B200 | 4× 180 GB720 GB total | 80% | $16.46 | $12,013 | UpCloudTrust 75 · read Nov 1 · receipt · also Deep Infra $19.56, Lambda $27.16 |
| H200 | 8× 141 GB1128 GB total | 51% | $21.04 | $15,359 |  Airon AInot scored · read Nov 1 · receipt · also Geodd $21.76, Vast.ai $22.11 |
| MI355X | 8× 288 GB2304 GB total | 25% | $23.60 | $17,228 |  TensorWaveTrust 64 · read Nov 1 · receipt |
| H200 NVL | 8× 141 GB1128 GB total | 51% | $26.09 | $19,042 |  IONOSnot scored · read Nov 1 · receipt · also Geodd $27.12, Vast.ai $39.47 |
| GB300 | 2× 288 GB576 GB total | 100% tight | $26.11 | $19,060 |  VerdaTrust 81 · read Nov 1 · receipt |
| B200 CC | 4× 180 GB720 GB total | 80% | $36.19 | $26,419 | VerdaTrust 81 · read Nov 1 · receipt |
| B300 CC | 4× 288 GB1152 GB total | 50% | $46.31 | $33,806 | VerdaTrust 81 · read Nov 1 · receipt |
| A800 80GB PCIe | 8× 80 GB640 GB total | 90% | $50.80 | $37,088 |  Volcano Enginenot scored · read Nov 1 · receipt |
| H100 NVL | 8× 94 GB752 GB total | 77% | $51.79 | $37,809 |  Open Telekom Cloudnot scored · read Nov 1 · receipt · also Cloud.ru $52.63 |
Throughput estimates · Oct 4, 2026Tokens per second, and what a million costs
Code. Speed estimated from memory bandwidth and model size, never measured. The cost per 1M output tokens is an estimate at the prices read on Oct 4, 2026; it is not recomputed here.
| GPU | Cards | Precision | tok/s per user | per 1M output tokens · estimate |
|---|---|---|---|---|
| MI300X | 4 | FP8 | 37.3 | $2.24 |
| B200 | 4 | FP8 | 65.7 | $2.92 |
| H100 SXM | 8 | FP8 | 40.8 | $3.76 |
| RTX PRO 6000 | 8 | FP8 | 24.4 | $5.68 |
| H100 PCIe | 8 | FP8 | 23.4 | $6.15 |
| H200 | 4 | FP8 | 48.0 | $7.37 |
cards, precision and speed estimated from memory bandwidth and model size; cost recomputed by the site from verified prices · cheapest per output at the prices of Oct 4, 2026: MI300XWhy estimates, and their limits →
Qwen3 Coder 480B A35B Instruct · related