freeqaz

decomp-synth-lifter-v14-evalformat-qwen3.5-9b-lora

README

License: apache-2.0

The project

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.

decomp-synth is the search-and-ML side of that system: an engine that searches behavior-preserving C++ rewrites against the real build, and mines its own verified attempts into training data. Where the single-turn siblings in this family propose blind, this adapter closes the loop: it is trained on the eval-format conversation shape, so the compiler's structured verdict on turn N is the prompt for turn N+1. The design rule stays the same throughout: models propose, deterministic oracles dispose — a fuzzy match score is a search gradient and an adversary, and byte-exact recompilation is the only terminal judge.

Background reading, in order:

How agent swarms decompile games byte-for-byte — the campaign, and what byte-exact matching actually demands
LLM-assisted decompilation: tools, not prompts — why the model is never allowed to judge its own work
Forking the decompilation toolchain — the compiler, splitter, and diff stack underneath
Infrastructure for running agent swarms — what running the fleet takes

Model facts

Table

base	`Qwen/Qwen3.5-9B` (apache-2.0)
method	LoRA r=64, α=128, dropout 0.05, bf16
corpus	3,946 chat-rendered rows: byte-exact repair chains (PowerPC X360 MSVC + GameCube/Wii mwcc), eval-format feedback panes
trained	2026-08-27
format	chat (Qwen chat template), single- and few-turn

Evaluation

Cell B conversation instrument: 4-turn compile-in-the-loop trajectories on 118 held-out start states (GameCube mwcc clone substrate), paired against a derangement-scrambled-feedback control, byte-exact witness as the endpoint. K=2 replicas, T=0.6.

vs the previous-generation v13 adapter: witnesses 10 vs 3 in both replicas (sign p=.039 / p=.016), pooled 20 vs 6 over 102 paired symbols (p=.022)
real-vs-scrambled feedback contrast resolves within one replica (+8 with zero reversals, p=.0078) — the gain is specifically in reading the feedback, not just solve count
caveats that travel with these numbers: unpreregistered capability A/B; the effect concentrates in start states already ≥97 fuzzy (near-done repairs); below ~90 the binding constraint for all 9B-class arms is compile success, not feedback reading

A follow-up rank ladder (r32/r64/r128 on a successor corpus) measured FLAT — adapter capacity is not the ceiling at 9B; data is.

How to use it

python
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-v14-evalformat-qwen3.5-9b-lora")

This adapter wants a loop, not a prompt. Unlike the RAW single-turn siblings, it is trained to converse with the compiler:

Turn 1: chat prompt carrying the target listing and the current C++ source (the project's eval-format layout).
Compile the model's proposal with the target project's real toolchain and score it with objdiff.
Turn 2+: append a structured feedback pane — fuzzy %, mismatching diff regions, compiler diagnostics — and let the model revise. The measured gain over v13 comes specifically from reading real feedback; scrambled feedback erases it.
Stop on a byte-exact witness, or after ~4 turns (the evaluated depth).

The feedback must be real: this model's edge is reading a true verdict, so wiring it to anything but an actual compile-and-score seam discards what it was trained for. It is strongest finishing near-matches (≥97 fuzzy) — pair it with a wide single-turn sampler (the v13/27B siblings) that gets functions close, then let this one land them. At 9B it fits a single 24 GB GPU in bf16; for serving, vLLM with --enable-lora works well.

The adapter family

Table with columns: adapter, base, format, role
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

Integrity

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.

Training data & license

Adapter weights: Apache-2.0. Training rows derive from public CC0-1.0-dedicated decompilation projects: Xbox 360 Rock Band 3 / Dance Central 3 community decompilations (MSVC PPC) and the zeldaret Wind Waker / Twilight Princess decompilations (mwcc GameCube). CC0 covers the decomp contributors' authorship, not the game publishers' rights in the underlying works; see the v10 card for the full discussion. Corpora are not published; PUBLISHED.json names the corpus file and sha256 as provenance.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

freeqaz

Model Tree

Base

Qwen/Qwen3.5-9B

Adapter

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality