Ornith-1.0-35B-FP8
Block-wise FP8 (E4M3) quantization of deepreinforce-ai/Ornith-1.0-35B — DeepReinforce's self-scaffolding agentic-coding MoE (Qwen3.5-35B-A3B hybrid). Served size ~35 GB; fits a single 80–96 GB GPU with full context.
Params
35.1 B
Context
—
Downloads 30d
193 K
Likes
8
Download history
daily snapshots · 91 days
▲ 0 in the last 30 days (0.0%)
193 K193 K
Oct 12Oct 22Nov 1Nov 10
863 K169 K
Aug 13Sep 12Oct 12Nov 10
863 K169 K
Aug 12Sep 11Oct 11Nov 10
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | f8_e4m3 | 37.6 GB | 47.1 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ curl -s https://aimodelscomparison.com/api/v1/models/ornith-1-0-35b-fp8-protolabsai
{
"hf_id": "protoLabsAI/Ornith-1.0-35B-FP8",
"params_b": 35.11,
"context_length": null,
"license": { "id": "mit", "commercial": "yes" },
"downloads_30d": 193326,
"vram_estimates": [
{ "quant": "f8_e4m3", "gb": 47.1 }
],
"updated_at": "2026-08-12T01:00:31Z"
}
Specifications
- Architecture
- Qwen3_5MoeForConditionalGeneration
- Parameters
- 35.1 B
- Tensor type
- F8_E4M3
- Licence
- mit
- First seen on the Hub
- 2026-06-27
- Base model
- Ornith-1.0-35B
- Training datasets
- undisclosed
- Added to our catalog
- 2026-08-12
Family
Base model and the most-downloaded derivatives in the catalog.
Compare with any text-generation model