Can I run Ornith 1.5 122B locally?
Ornith 1.5 122B by ornith-ai has 125B parameters (13B active per token, 256 experts, 8 used per token). At 4-bit with an 8K context it needs about 72.3 GB of GPU memory. It fits on 22 of 65 GPUs and Macs we track.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from unsloth/Ornith-1.5-122B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 44.7 GB | 44.7 GB | 45.3 GB | 47.6 GB |
| 3-bit (Q3_K_M) | usable | 57.9 GB | 58.0 GB | 58.6 GB | 60.8 GB |
| 4-bit (Q4_K_M) | good, most popular | 72.2 GB | 72.3 GB | 72.8 GB | 75.1 GB |
| 5-bit (Q5_K_M) | very good | 84.0 GB | 84.1 GB | 84.6 GB | 86.9 GB |
| 6-bit (Q6_K) | near lossless | 96.5 GB | 96.6 GB | 97.2 GB | 99.4 GB |
| 8-bit (Q8_0) • | effectively lossless | 124 GB | 125 GB | 125 GB | 127 GB |
| 16-bit (F16/BF16) • | original weights | 233 GB | 233 GB | 234 GB | 236 GB |
Which GPUs and Macs run Ornith 1.5 122B?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Questions
How much VRAM does Ornith 1.5 122B need?
About 72.3 GB at 4-bit (Q4_K_M) with an 8K context: 71.2 GB of weights (estimated), 0.2 GB of KV cache and 0.9 GB of runtime overhead. At 8-bit it needs about 125 GB.
What is the smallest GPU that runs Ornith 1.5 122B?
No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the RTX PRO 6000 Blackwell (96 GB).
Can I run Ornith 1.5 122B on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M3 Max 128GB or larger, using the default macOS GPU memory limit.
Similar models
Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.
Ornith 1.5 35B A3B
36B · 21.4 GB at 4-bitOrnith 1.5 9B
9.7B · 6.6 GB at 4-bitLing 3.0 flash Sante
127B · 75.9 GB at 4-bitLing 3.0 flash
127B · 73.5 GB at 4-bitgpt oss 120b
117B · 59.5 GB at 4-bitgpt oss safeguard 120b
117B · 59.5 GB at 4-bitModel facts
| Hugging Face repo | ornith-ai/Ornith-1.5-122B |
| Parameters | 125,086,497,008 (125B) · ~13B active (computed from config) |
| Layers / hidden size | 48 / 3072 |
| Attention | GQA (hybrid: 36 linear/state layers); 32 heads, 2 KV heads, head dim 256 |
| KV cache per token (FP16) | 24.0 KiB |
| Max context | 262,144 tokens |
| License | mit |
| Released on HF | Oct 16, 2026 |
| Downloads (30d) / likes | 359,058 / 227 |
Sources: Hugging Face model API · config.json (ornith-ai/Ornith-1.5-122B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.