Every claim on this card is verified
Trust: self-attested · 2 benchmarks · 1 device tested
ForgeAlloy chain of custody ·
Download alloy · Merkle-chained
The first 30B-class coder that fits a 12 GB consumer GPU. Forged from
Qwen/Qwen3-Coder-30B-A3B-Instruct by removing 35% of MoE experts via
calibration-aware activation-frequency ranking on a held-out Python code corpus. Quantized to GGUF Q4_K_M / Q5_K_M / Q8_0 for llama.cpp / Ollama / LM Studio. Native
256K context preserved (the only Qwen3-Coder variant in this VRAM tier with 256K). Apache-2.0.
88.4 HumanEval against the unmodified base's
92.1 (Δ −3.7), both hardware-measured on RTX 5090 in the same eval pipeline. This is the only Qwen3-Coder-class artifact at the 12 GB tier. Cryptographic provenance via
ForgeAlloy — the published per-problem JSONL outputs let anyone re-score against the alloy's resultHash without trusting the producer's claim.
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model = AutoModelForCausalLM.from_pretrained("continuum-ai/qwen3-coder-30b-a3b-compacted-19b-256k",
4 torch_dtype="auto", device_map="auto")
5tokenizer = AutoTokenizer.from_pretrained("continuum-ai/qwen3-coder-30b-a3b-compacted-19b-256k")
6
7inputs = tokenizer("def merge_sort(arr):", return_tensors="pt").to(model.device)
8output = model.generate(**inputs, max_new_tokens=200)
9print(tokenizer.decode(output[0], skip_special_tokens=True))
Produced via MoE expert pruning, GGUF quantization. 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
expert-activation-profile → expert-prune → quant → eval over 1 cycle on NVIDIA GeForce RTX 5090.
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.
