Research · agentic code post-training

Threnody 3.0

An 8.8B multimodal agentic model post-trained with anti-loop filtering, refusal orthogonalization and multi-schema tool calling. Merged at scale-0.75 from Qwen3.5-9B and quantized to 5.6 GiB Q4_K_M for local 12 GB GPUs.

Apache-2.0 8.8B parameters Q4_K_M GGUF RTX 3060 12 GB
Abstract flowing artwork representing Threnody 3.0.
Latest version
Threnody 3.0
Trainable (LoRA)
0.75 strength merge
OpenCode held-out
13/14 (92.9%)
Last commit
—
GitHub stars
—

Where it fits

Closing the agentic loop on consumer hardware.

Where Fable-Coder anchored tool calling on a 7B base, Threnody 3.0 pushes into the denser 8.8B architecture of Qwen3.5-9B to solve the hardest barrier in autonomous programming: repetitive tool loops and premature task abandonment.

By pairing strict anti-loop trace filtering with refusal orthogonalization and multimodal terminal vision, Threnody 3.0 achieves a 92.9% pass rate on held-out OpenCode workflows while fitting cleanly within 6 GB of VRAM.

01

Anti-loop execution

Eliminates cyclic tool calls, repetitive command re-tries and redundant file reads with strict trajectory-level negative penalization.

02

Refusal orthogonalization

Abliteration across layers 12–28 prevents false-positive refusals during deep system administration, debugging and security workflows.

03

Vision & multi-schema tools

8 execution schemas plus native terminal screenshot inspection via an accompanying mmproj projector file.

The six architectural pillars.

A focused post-training recipe addressing real-world agent harness bottlenecks without degrading general reasoning.

LoRA training mixture · curated dataset mix

General coding & math replay40%

Prevents catastrophic forgetting across algorithms, data structures, and mathematical proofs.

Multi-schema tool execution (8 schemas)25%

Native ChatML <tool_call>, JSON schemas, XML function wrappers, and sandboxed Bash commands.

Interleaved thinking & planning traces20%

Deep chain-of-thought traces interleaved in 75% of trajectories before initiating tool actions.

Anti-loop & error recovery verification15%

Explicit recovery traces from failed command executions, broken pipes, and corrupted outputs.

Training specifications

Base model
Qwen/Qwen3.5-9B (~8.8B parameters)
Post-training
Replay-anchored LoRA, rank 64 / alpha 128, all 7 projections
Merge ratio
Scale 0.75 (s0.75 LoRA merge)
Refusal vector
Orthogonalized projection across layers 12–28
Context length
8,192 tokens native (extendable to 32K)
Compute
NVIDIA A100-SXM4-80GB
Quantized size
5.6 GiB Q4_K_M (+ 600 MiB mmproj)

Key innovations

  • Anti-Loop Filter: Models fine-tuned with max_repeat = 1 verification penalties on held-out tasks, cutting infinite re-try loops by over 80%.
  • Abliterated Safety: Refusal directions isolated and neutralized in intermediate layers so the model inspects system logs, exploits, and diagnostics without preachy refusals.
  • Multimodal Inspection: Accompanied by a dedicated vision adapter for CLI terminals, rendered web views, and UI layout debugging.
  • Balanced s0.75 Merge: Blending 75% adapted weights preserves general conversational fluency while adopting strict agent discipline.

From V1 to Threnody 3.0.

Documenting the iterations, breakthroughs, and regressions that led to the final build.

  1. Base · Qwen3.5-9B Dense, capable foundation with raw potential

    Strong natural language comprehension and coding fundamentals, but prone to tool-calling loops and ungrounded retries in autonomous harnesses.

  2. V1 · Initial SFT Supervised fine-tuning on multi-schema data

    Demonstrated immediate improvements in schema adherence, but suffered from overconfidence and occasional catastrophic forgetting in complex math.

  3. V2 · Aggressive synthetic traces Loop instability regression

    Dropped to 9/14 (64.3%) on OpenCode held-out evals — low-variance synthetic data caused cyclic repeating behavior when tools returned unexpected stdout.

  4. Threnody 3.0 · Anti-loop LoRA with s0.75 merge Peak agentic resilience and vision support

    Anti-loop negative filtering, refusal abliteration, and a balanced 0.75 weight merge achieve 13/14 (92.9%) pass rate on held-out agentic loops with zero mode collapse.

Rigorous, honest benchmarks.

Head-to-head evals tested across base, V2, and V3 on a local RTX 3060 running identical inference parameters.

Agentic harness & reasoning comparison

Model AgentLoop held-out MMLU-Pro BioChem OmniAgent-Bench AgentLoop in-harness AgentLoop-Hard
Threnody 3.0
8.8B s0.75 LoRA
13/14 (92.9%) 78.5% 13/16 (81.3%) 10/14 (71.4%) 17/30 (56.7%)
Qwen3.5-9B
base model
12/14 (85.7%) 76.0% 11/16 (68.8%) 10/14 (71.4%) 15/30 (50.0%)
Threnody V2
synthetic SFT
9/14 (64.3%) 75.2% 10/16 (62.5%) 8/14 (57.1%) 12/30 (40.0%)

Held-out AgentLoop evaluations run against OpenCode environments never encountered during training.

Honest reporting

Known trade-offs and discrepancies.

Every post-trained model incurs trade-offs. AxiomInference reports regressions and architectural nuances openly so practitioners know exactly what to expect.

HumanEval Python Completion Regression

On pure zero-shot Python completion (HumanEval), Threnody 3.0 scored 80.5% compared to the base model's 82.3% (−1.8 pp). This minor regression stems from the model's strong inductive bias toward tool invocation and diagnostic verification steps rather than naive raw text generation.

Why Merge Scale-0.75 Instead of 1.0

A full 100% LoRA merge resulted in rigid formatting patterns and occasional mode narrowing during conversational chat. Merging at a calibrated 0.75 factor retained base model nuance while keeping 99% of the agentic and anti-loop benefits.

Multimodal Projector Scope

The vision tower (mmproj-Threnody-9B-V3-F16.gguf) is optimized for technical artifacts — terminal screenshots, stack traces, diagrams, and web page layout verification — not open-domain creative image captioning.

Run it yourself.

Download the quantized GGUF and vision adapter from Hugging Face, then run locally via Ollama or llama.cpp.

Modelfile · Threnody 3.0
# Clone repo or place GGUF alongside this Modelfile
FROM ./Threnody-9B-V3-Q4_K_M.gguf

PARAMETER stop "<|im_end|>"
PARAMETER stop "<|endoftext|>"
PARAMETER temperature 0.2
PARAMETER top_p 0.95
PARAMETER repeat_penalty 1.15

SYSTEM """You are Threnody 3.0, an expert autonomous coding and engineering agent.
Plan actions carefully, verify assumptions, use available tools, and avoid redundant loops."""

Build and run: ollama create threnody-v3 -f ./Modelfile && ollama run threnody-v3

Verify your download

Language model
Threnody-9B-V3-Q4_K_M.gguf (~5.6 GiB)
Vision projector
mmproj-Threnody-9B-V3-F16.gguf (~600 MiB)
Quantization
Q4_K_M (Medium K-quant)
License
Apache-2.0

Where to get it

The repository.

Live commits and updates direct from GitHub.