Downloads · 30 days
5
11% of all-time downloads
freeqaz/decomp-synth-lifter-v17-full-qwen3.5-9b-r64-lora
decomp-synth-lifter-v17-full-qwen3.5-9b-r64-lora is a machine learning model from freeqaz. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.
The r=64 twin of v17-full r32: trained on the byte-identical corpus (v17full.jsonl, sha 411cfaed…, 3,032 tool-trajectory rows) with the same schedule, step count, and seed, and LoRA rank and α doubled (r=64, α=128, sc…
Downloads · 30 days
5
11% of all-time downloads
All-time downloads
45
Public
Repo size
713 MB
Likes
0
Public
Click a slice to open those files.
.safetensors693 MB · 97%
From the Hugging Face model README
The r=64 twin of
v17-full r32:
trained on the byte-identical corpus (v17full.jsonl, sha 411cfaed…,
3,032 tool-trajectory rows) with the same schedule, step count, and seed, and
LoRA rank and α doubled (r=64, α=128, scaling held at α/r=2.0). Everything
else is fixed — the pair isolates a single question: does doubling adapter
capacity matter on this supply?
Like its r32 sibling, this is the family's first multi-turn tool-use
trajectory generation: a teacher agent (GLM-5.3-flash) driving the
project's real compile-and-diff tools over decompilation targets, its
sessions converted to training rows and rendered through Qwen3.5-9B's own
chat template — teacher reasoning in the native <think> channel, tool
calls in the template's XML form.
Freshly trained; not yet evaluated — see the status notes below.
This adapter comes out of a campaign teaching fleets of AI coding agents to reconstruct shipped game binaries byte-for-byte — matching decompilation, the console-preservation community's gold standard: rewrite C++ until the original vintage compiler, at the original flags, reproduces the retail machine code exactly. In that campaign, agent swarms byte-matched over 44,000 functions of a stripped retail Xbox 360 binary in about ten weeks — with every proposal judged by the real compiler and objdiff, never by a model. Start with the flagship write-up and the methodology post; the full project context lives on the r32 card, which is the primary release of this pair.
The family has run this experiment before: v13-chain r64 vs r32 was a measured null — doubled adapter capacity at fixed data bought nothing detectable at 9B — and a later internal rank ladder (v16) was flat as well. That history is exactly why the r32 arms are this generation's primaries. But v17's supply is qualitatively different from anything those rank experiments trained on (multi-turn trajectories to ~87k tokens, tool calls, recovery sequences), so the project re-ran the twin rather than assuming the null transfers.
No verdict exists yet: unlike the v13 pair, neither v17 arm has been
evaluated, so this pair currently offers matched artifacts, not a
measured contrast. Both repos' training-job-written PUBLISHED.json pin the
same corpus sha256, so the corpus difference is provably zero for anyone who
wants to run the comparison.
Published as a research artifact with its evidence stated plainly:
v17-matched arm.| base | Qwen/Qwen3.5-9B (apache-2.0) |
| method | LoRA r=64, α=128, dropout 0.05, all-linear, bf16 |
| corpus | 3,032 tool-trajectory rows (v17full.jsonl, sha 411cfaed…) — byte-identical to the r32 arm |
| window | 90,112 tokens max_seq (zero rows dropped) |
| schedule | 2 epochs, 190 steps, eff. batch 32, linear LR, seed 42 |
| trained | 2026-08-31 |
| format | Qwen3.5 chat template, thinking ON, native XML tool calls |
The corpus is the project's 2026-08-30/31 teacher-mint union: multi-turn trajectories where a tool-armed teacher (GLM-5.3-flash) works decompilation targets from a private Halo: Combat Evolved Anniversary (Xbox 360) reconstruction effort through the project's compile-and-score harness — reading disassembly, proposing C++, compiling with the real toolchain, and reacting to objdiff feedback across turns (up to 8). Trajectories are converted at tool-turn granularity, holdout-gated at translation-unit grain against the project's eval sets, and student-rendered through Qwen3.5-9B's own tokenizer template.
The completions are therefore teacher-model-generated text plus tool-call
structure; the surrounding context (assembly, diffs, compiler output)
derives from that private reconstruction effort, which is not publicly
linkable. No license claim is made for the corpus; it is not published. The
shipped PUBLISHED.json pins the corpus file's sha256
(411cfaedcae6…, 3,032 rows) as provenance.
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-9B", dtype="bfloat16", device_map="auto")
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-9B")
model = PeftModel.from_pretrained(base, "freeqaz/decomp-synth-lifter-v17-full-qwen3.5-9b-r64-lora")
This adapter is trained for agentic tool use: serve it behind an OpenAI-compatible endpoint with tool calling enabled and give it compile / diff / read tools. With vLLM:
vllm serve Qwen/Qwen3.5-9B --enable-lora \
--lora-modules v17r64=freeqaz/decomp-synth-lifter-v17-full-qwen3.5-9b-r64-lora \
--enable-auto-tool-choice --tool-call-parser qwen3_coder
The tool-call parser flag matters: the adapter emits tool calls in the
Qwen3.5 template's native XML form, which qwen3_coder parses; without it
(or with a JSON-style parser) the calls degrade silently into plain text.
Like every adapter in this family, it is a proposal engine for a compile-and-score loop — sample multiple candidates, compile each with the target project's real toolchain, and let objdiff judge. Byte-exact is the only acceptance criterion. Training rows ran up to ~87k tokens of session context, so give it a long context window. In bf16 the 9B base fits a single 24 GB GPU.
| adapter | base | format | role |
|---|---|---|---|
| v10 | Qwen3.6-27B | RAW | single-turn matcher, reference 27B |
| gtp0c | Qwen3.8-27B | RAW | v10 recipe verbatim on the newer base (matched pair) |
| v13-chain r32 | Qwen3.5-9B | RAW | headline single-turn 9B |
| v13-chain r64 | Qwen3.5-9B | RAW | rank-capacity null twin of the r32 |
| v14-evalformat | Qwen3.5-9B | chat | multi-turn feedback-driven, strongest measured agentic 9B |
| v16-v5chat | Qwen3.5-9B | chat | broadest corpus, research artifact with stated negatives |
| v17-full r32 | Qwen3.5-9B | chat+tools | first tool-trajectory generation, unevaluated |
| v17-matched | Qwen3.5-9B | chat+tools | byte-exact-only trajectory filter of v17-full (contrast arm), unevaluated |
| v17-full r64 (this repo) | Qwen3.5-9B | chat+tools | rank-capacity twin of v17-full r32 (r64/α128), unevaluated |
| v17-mix | Qwen3.5-9B | chat+tools | v17 trajectories + single-turn supply mix (broadest v17 corpus), unevaluated |
PUBLISHED.json (shipped) carries per-file sha256 written by the training
job at publish time. Two internal metadata files it pins
(train_summary.json, artifact-manifest.json) are retained in the
project's artifact store but not shipped here — they carry training-box
paths. adapter_config.json was rewritten post-training to the HF base id;
original bytes at provenance/adapter_config.orig.json.
Adapter weights: Apache-2.0. A model trained to reproduce shipped game code can reproduce shipped game code — that is its purpose; decompilation projects operate in a long-standing community norm of clean-room reconstruction for interoperability and preservation.