Ornith 1 models
17 Ornith 1 models from ornith-ai on Hugging Face, from 4.3B to 403.4B parameters, published by ornith-ai in BF16, with a 256K-token context. The smallest official model, Ornith-1.5-9B, needs about 21.6 GB of VRAM at its published precision; the cheapest live fit is RTX A5000 at $0.176/hr.
Pick a size
One row per official Ornith 1 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| Ornith-1.5-9B | 9.7B | 21.6 GB | 10.8 GB | 5.4 GB | RTX A5000 | $0.176/hr | 1.0 GB |
| Ornith-1.5-35B-A3B | 36.0B | 80.4 GB | 40.2 GB | 20.1 GB | RTX PRO 6000 | $1.38/hr | 0.63 GB |
| Ornith-1.0-397B | 396.8B | 887 GB | 443 GB | 222 GB | AMD MI300X × 5 | $13.14/hr | 0.94 GB |
| Ornith-1.5-397B | 403.4B | 902 GB | 451 GB | 225 GB | AMD MI300X × 5 | $13.14/hr | 0.94 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Ornith-1.5-9B | 9.7B | BF16 | 21.6 GB | RTX A5000 | $0.176/hr |
| Ornith-1.5-35B-A3B | 36.0B | BF16 | 80.4 GB | RTX PRO 6000 | $1.38/hr |
| Ornith-1.0-397B | 396.8B | BF16 | 887 GB | AMD MI300X × 5 | $13.14/hr |
| Ornith-1.5-397B | 403.4B | BF16 | 902 GB | AMD MI300X × 5 | $13.14/hr |
Quantized versions (11)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Ornith-1.5-9B-INT4-W4A16-AutoRound | 4.3B | BF16 | 9.7 GB | RTX 3060 | $0.110/hr |
| Ornith-1.5-9B-MLX | 9.0B | BF16 | 20.0 GB | RTX A5000 | $0.176/hr |
| Ornith-1.5-35B-A3B-MLX | 34.7B | BF16 | 77.5 GB | A100 | $1.18/hr |
| Ornith-1.0-35B-FP8 | 35.1B | F8_E4M3 | 39.2 GB | L40 | $0.759/hr |
| Ornith-1.0-35B-FP8 | 35.1B | F8_E4M3 | 39.2 GB | L40 | $0.759/hr |
| Ornith-1.0-35B-AWQ-FP8 | 35.1B | F8_E4M3 | 39.3 GB | L40 | $0.759/hr |
| Ornith-1.0-35B-FP8 | 35.1B | F8_E4M3 | 39.3 GB | L40 | $0.759/hr |
| Ornith-1.5-35B-A3B-FP8 | 36.0B | F8_E4M3 | 40.2 GB | L40 | $0.759/hr |
| Ornith-1.0-397B-FP8 | 396.8B | F8_E4M3 | 443 GB | RTX PRO 6000 × 5 | $6.88/hr |
| Ornith-1.0-397B-FP8 | 397.0B | F8_E4M3 | 444 GB | RTX PRO 6000 × 5 | $6.88/hr |
| Ornith-1.5-397B-FP8 | 403.4B | F8_E4M3 | 451 GB | RTX PRO 6000 × 5 | $6.88/hr |
Fine-tunes and community models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Ornith-1.5-9B-OBLITERATED | 9.7B | BF16 | 21.6 GB | RTX A5000 | $0.176/hr |
| Ornith-1.0-35B-AEON-Ultimate-Uncensored-BF16 | 35.1B | BF16 | 78.5 GB | A100 | $1.18/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.