Ideogram · Image generation · 9.3B parameters · Ideogram 4 Non-Commercial
Ideogram 4
Needs 12 GB at FP8 / INT8. The smallest real configuration today is 1× RTX 3060 at Vast.ai, $0.04 per hour, about $26 a month at 730 hours.
Parameters9.3Bdense
CategoryImage generationsized on weights + headroom
Checkpoint28 GBFP8, as published
LicenseIdeogram 4 Non-Commercialcheck before commercial use
ReleasedJun 3, 2026documented date
Smallest configuration · FP8 / INT8
1× RTX 306012 GB VRAM · 93% used
- Per hour
- $0.04
- Per month
- $26
- Cheapest at
Vast.ai
What is documented about Ideogram 4
news checked 2026-10-27Ideogram's first open-weight model, trained from scratch. Files: transformer 9.29 GB FP8 + separate 'unconditional' transformer 9.29 GB (dual-branch CFG) + text encoder 8.78 GB (Qwen3-VL-8B-Instruct, 13 layers of hidden states) + VAE 0.17 GB; NF4 repo 16.1 GB. NOT in the 2026-10-04 catalogue although Comfy-Org packaged it (180K downloads). Ideogram 4.5 / Vector 1.0 are closed.
- https://huggingface.co/ideogram-ai/ideogram-4-fp8
- https://huggingface.co/api/models/ideogram-ai/ideogram-4-fp8?blobs=true
- https://huggingface.co/api/models/ideogram-ai/ideogram-4-nf4?blobs=true
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 |
|---|---|---|
| FP8published checkpoint | 28 GB | 33 GB |
| FP8 / INT8estimated | 9 GB | 12 GB |
| INT4 / MXFP4estimated | 6 GB | 7 GB |
Cross-checked on Hugging Face Oct 28, 2026: 9.28B parameters in the safetensors headers, 27.5 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 |
|---|---|---|---|---|---|
| RTX 3060Best fit | 1× 12 GB12 GB total | 93% tight | $0.04 | $26 | Vast.aiTrust 78 · read Nov 1 · receipt · also Clore.ai $0.05, gcube $0.05 |
| RTX A2000 | 1× 12 GB12 GB total | 93% tight | $0.04 | $26 | Vast.aiTrust 78 · read Nov 1 · receipt · also gcube $0.09 |
| T4 | 1× 16 GB16 GB total | 70% | $0.04 | $29 | Vast.aiTrust 78 · read Nov 1 · receipt · also iwinv $0.20, Theta EdgeCloud $0.20 |
| Tesla P40 | 1× 24 GB24 GB total | 47% | $0.04 | $30 |  Clore.aiTrust 46 · read Nov 1 · receipt · also gcube $0.07, Vast.ai $0.11 |
| GTX Titan X | 1× 12 GB12 GB total | 93% tight | $0.04 | $31 | Vast.aiTrust 78 · read Nov 1 · receipt |
| Titan Xp | 1× 12 GB12 GB total | 93% tight | $0.04 | $31 | Vast.aiTrust 78 · read Nov 1 · receipt |
| Tesla P100 | 1× 16 GB16 GB total | 70% | $0.04 | $33 | Vast.aiTrust 78 · read Nov 1 · receipt · also gcube $0.07 |
| RTX A4000 | 1× 16 GB16 GB total | 70% | $0.06 | $46 | Clore.aiTrust 46 · read Nov 1 · receipt · also Vast.ai $0.07, Edgevana $0.10 |
| Tesla V100 16GB | 1× 16 GB16 GB total | 70% | $0.06 | $46 | Clore.aiTrust 46 · read Nov 1 · receipt · also Vast.ai $0.08, CloudRift $0.28 |
| Tesla V100 32GB | 1× 32 GB32 GB total | 35% | $0.07 | $50 | Vast.aiTrust 78 · read Nov 1 · receipt · also Cyfuture AI $0.57, E2E Networks $1.50 |
| RTX PRO 2000 | 1× 16 GB16 GB total | 70% | $0.07 | $51 | Clore.aiTrust 46 · read Nov 1 · receipt · also gcube $0.16 |
| Tesla M40 | 1× 24 GB24 GB total | 47% | $0.07 | $54 | Clore.aiTrust 46 · read Nov 1 · receipt · also gcube $0.07 |
| RTX 3080 Ti | 1× 12 GB12 GB total | 93% tight | $0.08 | $60 | Vast.aiTrust 78 · read Nov 1 · receipt · also Clore.ai $0.10, gcube $0.10 |
| RTX 4070 | 1× 12 GB12 GB total | 93% tight | $0.08 | $61 | Clore.aiTrust 46 · read Nov 1 · receipt · also Vast.ai $0.12, gcube $0.14 |
| GTX 1660 | 2× 6 GB12 GB total | 93% tight | $0.09 | $64 | Vast.aiTrust 78 · read Nov 1 · receipt |
| RTX 5060 Ti | 1× 16 GB16 GB total | 70% | $0.09 | $64 | Clore.aiTrust 46 · read Nov 1 · receipt · also Vast.ai $0.09, Theta EdgeCloud $0.12 |
| RTX 3090 | 1× 24 GB24 GB total | 47% | $0.09 | $66 | Akash NetworkTrust 53 · read Nov 1 · receipt · also Vast.ai $0.10, Clore.ai $0.11 |
| RTX 4070 Super | 1× 12 GB12 GB total | 93% tight | $0.10 | $70 | Vast.aiTrust 78 · read Nov 1 · receipt · also gcube $0.15, Clore.ai $0.17 |
| GTX 1070 | 2× 8 GB16 GB total | 70% | $0.10 | $70 | Vast.aiTrust 78 · read Nov 1 · receipt |
| GTX 1070 Ti | 2× 8 GB16 GB total | 70% | $0.10 | $76 | Vast.aiTrust 78 · read Nov 1 · receipt |
| L4 | 1× 24 GB24 GB total | 47% | $0.10 | $76 | Vast.aiTrust 78 · read Nov 1 · receipt · also Hikube $0.27, Infomaniak $0.28 |
| GTX 1080 | 2× 8 GB16 GB total | 70% | $0.11 | $79 | Vast.aiTrust 78 · read Nov 1 · receipt |
| RTX 2070 | 2× 8 GB16 GB total | 70% | $0.11 | $79 | Vast.aiTrust 78 · read Nov 1 · receipt · also Edgevana $0.14 |
| Quadro P2000 | 4× 5 GB20 GB total | 56% | $0.11 | $80 | Vast.aiTrust 78 · read Nov 1 · receipt |
| The 24 cheapest of 100 GPU types that fit. Every GPU, any precision and context, in the finder | |||||
Ideogram 4 · related