Tencent · Image generation · 83B parameters · Tencent Hunyuan Community License
HunyuanImage 3.0
Needs 100 GB at FP8 / INT8. The smallest real configuration today is 4× Tesla V100 32GB at Vast.ai, $0.27 per hour, about $196 a month at 730 hours.
Parameters83Bdense
CategoryImage generationsized on weights + headroom
Checkpoint169 GBBF16, as published
LicenseTencent Hunyuan Community Licensecheck before commercial use
ReleasedSep 28, 2025documented date
Smallest configuration · FP8 / INT8
4× Tesla V100 32GB128 GB VRAM · 78% used
- Per hour
- $0.27
- Per month
- $196
- Cheapest at
Vast.ai
Memory · 3 precisionsHow much VRAM it needs
Weights plus 20% runtime overhead. For image generation models, activations also depend on resolution, clip length and batch size; the per-GPU estimates below give the practical configuration.
| Precision | Weights | Weights + headroom |
|---|---|---|
| BF16published checkpoint | 169 GB | 203 GB |
| FP8 / INT8estimated | 83 GB | 100 GB |
| INT4 / MXFP4estimated | 50 GB | 60 GB |
Cross-checked on Hugging Face Oct 28, 2026: 83.0B parameters in the safetensors headers, 168.5 GB of weight files. 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 not added (architecture not readable).
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 |
|---|---|---|---|---|---|
| Tesla V100 32GBBest fit | 4× 32 GB128 GB total | 78% | $0.27 | $196 | Vast.aiTrust 78 · read Nov 1 · receipt · also Clore.ai $0.41, Cyfuture AI $2.22 |
| Tesla P100 | 8× 16 GB128 GB total | 78% | $0.38 | $274 | Vast.aiTrust 78 · read Nov 1 · receipt |
| Tesla V100 16GB | 8× 16 GB128 GB total | 78% | $0.75 | $547 | Vast.aiTrust 78 · read Nov 1 · receipt · also CloudRift $2.32, Lambda $6.32 |
| RTX 5060 Ti | 8× 16 GB128 GB total | 78% | $0.75 | $549 | Vast.aiTrust 78 · read Nov 1 · receipt · also Clore.ai $0.82, Edgevana $1.51 |
| RTX A4000 | 8× 16 GB128 GB total | 78% | $0.96 | $701 |  LiumTrust 40 · read Nov 1 · receipt · also Vast.ai $0.96, VoltageGPU $1.92 |
| A800 80GB PCIe | 2× 80 GB160 GB total | 62% | $0.96 | $702 | Vast.aiTrust 78 · read Nov 1 · receipt |
| Quadro RTX 8000 | 4× 48 GB192 GB total | 52% | $1.01 | $741 | Vast.aiTrust 78 · read Nov 1 · receipt |
| RTX 3090 | 8× 24 GB192 GB total | 52% | $1.17 | $851 |  Clore.aiTrust 46 · read Nov 1 · receipt · also Vast.ai $1.29, Lium $2.32 |
| A100 80GB | 2× 80 GB160 GB total | 62% | $1.20 | $878 | Vast.aiTrust 78 · read Nov 1 · receipt · also Thunder Compute $2.18, Spheron $2.38 |
| L40S | 4× 48 GB192 GB total | 52% | $1.52 | $1,110 | LiumTrust 40 · read Nov 1 · receipt · also Vast.ai $1.87, VoltageGPU $3.04 |
| A100 40GB | 4× 40 GB160 GB total | 62% | $1.60 | $1,170 | Vast.aiTrust 78 · read Nov 1 · receipt · also iwinv $3.66, AtmosCompute $4.05 |
| RTX 5090 | 4× 32 GB128 GB total | 78% | $1.60 | $1,170 | Vast.aiTrust 78 · read Nov 1 · receipt · also Clore.ai $2.33, Edgevana $3.31 |
| MI300X | 1× 192 GB192 GB total | 52% | $1.65 | $1,205 |  GPUBankTrust 93 · read Oct 25 · receipt · also OpenRelay $1.99, Runpod $2.39 |
| RTX PRO 6000 | 2× 96 GB192 GB total | 52% | $1.67 | $1,218 | Vast.aiTrust 78 · read Nov 1 · receipt · also UpCloud $1.94, Edgevana $3.07 |
| RTX A6000 | 4× 48 GB192 GB total | 52% | $1.68 | $1,226 | LiumTrust 40 · read Nov 1 · receipt · also Spheron $1.77, GPU.ai $1.88 |
| RTX 5080 | 8× 16 GB128 GB total | 78% | $1.82 | $1,328 | Vast.aiTrust 78 · read Nov 1 · receipt · also Edgevana $2.34, AtmosCompute $3.27 |
| RTX 5070 Ti | 8× 16 GB128 GB total | 78% | $1.82 | $1,331 | Vast.aiTrust 78 · read Nov 1 · receipt · also Clore.ai $1.88, Edgevana $2.10 |
| RTX PRO 5000 | 4× 48 GB192 GB total | 52% | $1.92 | $1,403 | Vast.aiTrust 78 · read Nov 1 · receipt · also Edgevana $4.60, AtmosCompute $5.28 |
| RTX PRO 4000 | 8× 24 GB192 GB total | 52% | $1.93 | $1,406 | Vast.aiTrust 78 · read Nov 1 · receipt · also Edgevana $2.15, AtmosCompute $3.47 |
| RTX 4070 Ti Super | 8× 16 GB128 GB total | 78% | $2.08 | $1,521 | Clore.aiTrust 46 · read Nov 1 · receipt |
| H200 | 1× 141 GB141 GB total | 71% | $2.09 | $1,526 |  BeamTrust 76 · read Nov 1 · receipt · also GPUBank $2.49, Seeweb $2.90 |
| Gaudi 2 | 2× 96 GB192 GB total | 52% | $2.11 | $1,538 |  Cyfuture AInot scored · read Nov 1 · receipt |
| RTX 4000 Ada | 8× 20 GB160 GB total | 62% | $2.14 | $1,559 | Vast.aiTrust 78 · read Nov 1 · receipt |
| MI325X | 1× 256 GB256 GB total | 39% | $2.25 | $1,643 |  OpenRelaynot scored · read Nov 1 · receipt · also Cyfuture AI $3.14, DigitalOcean $3.80 |
| The 24 cheapest of 49 GPU types that fit. Every GPU, any precision and context, in the finder | |||||
Throughput estimates · Oct 4, 2026Estimated speed, in s per image
Image generation. Speed estimated from memory bandwidth and model size, never measured. The cost per 1,000 images is an estimate at the prices read on Oct 4, 2026; it is not recomputed here.
| GPU | Cards | Precision | s per image | per 1,000 images · estimate |
|---|---|---|---|---|
| RTX PRO 6000 | 2 | FP8 | 5.4 | $3.46 |
| B200 | 1 | FP8 | 3.1 | $3.60 |
| H100 SXM | 2 | FP8 | 3.8 | $3.85 |
| L40S | 4 | FP8 | 6.5 | $3.97 |
| H200 | 1 | FP8 | 5.8 | $4.05 |
| MI300X | 1 | FP8 | 9.0 | $4.11 |
| H100 PCIe | 2 | FP8 | 6.0 | $4.47 |
| A100 80GB | 4 | BF16 | 4.8 | $5.05 |
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: RTX PRO 6000Why estimates, and their limits →
HunyuanImage 3.0 · related