30% Smaller, -3.2% Better
Qwen2.5-0.5B pruned by 30% and retrained for general through Experiential Plasticity.
2.83 → 2.92 perplexity · 3 cycles
Every claim on this card is verified
Trust: self-attested · 1 benchmark · 2 devices tested
ForgeAlloy chain of custody · Download alloy · Merkle-chained
Qwen2.5-0.5B with cryptographic provenance via the ForgeAlloy chain of custody.
Benchmarks
| Benchmark | Result | Verified |
|---|---|---|
| perplexity | 2.9 | Self-reported |
What Changed (Base → Forged)
| Base | Forged | Delta | |
|---|---|---|---|
| Perplexity (general) | 2.83 | 2.92 | +3.2% ✅ |
| Pruning | None | 30% heads (magnitude) | -30% params ✅ |
| Training | General | general, 1000 steps | LR 2e-4, 3 cycles |
| Pipeline | prune → train | 3 cycles |
Runs On
| Device | Format | Size | Speed |
|---|---|---|---|
| Phone / Raspberry Pi | Q4_K_M | — | Verified |
| MacBook Air 8GB | Q4_K_M | — | Verified |
| MacBook Pro 32GB | fp16 | 8.0GB | Expected |
| MacBook Air 16GB | Q8_0 | ~4.0GB | Expected |
| MacBook Air 8GB | Q4_K_M | ~2.5GB | Expected |
| iPhone / Android | Q4_K_M | ~2.5GB | Expected |
Quick Start
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("continuum-ai/qwen2.5-0.5b-general-forged",
torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("continuum-ai/qwen2.5-0.5b-general-forged")
inputs = tokenizer("def merge_sort(arr):", return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(output[0], skip_special_tokens=True))
Methodology
Produced via head pruning. Full methodology, ablations, and per-stage rationale are in the methodology paper and the companion MODEL_METHODOLOGY.md in this repository. The pipeline ran as prune → train over 3 cycles on Phone / Raspberry Pi.
Chain of Custody
Scan the QR or verify online. Download the alloy file to verify independently.
| What | Proof |
|---|---|
| Model weights | sha256:3f1fd0b9b5b70d7869fd9a48162641f76... |
| Code that ran | sha256:legacy-pre-alloy-... |
| Forged on | Phone / Raspberry Pi, 2026-03-27T08:22:03-05:00 |
| Trust level | self-attested |
| Spec | ForgeAlloy — Rust/Python/TypeScript |
Make Your Own
Forged with Continuum — a distributed AI world that runs on your hardware.
The Factory configurator lets you design and forge custom models visually — context extension, pruning, LoRA, quantization, vision/audio modalities. Pick your target devices, the system figures out what fits.
GitHub · All Models · Forge-Alloy
License
apache-2.0