Tag

lordx64/agentic-distill-fable-5-sft

Provides 4,659 agentic single-turn SFT training pairs extracted from Claude Fable‑5, formatted as a single-column parquet for Qwen-style fine-tuning. Includes explicit chain-of-thought (<think>) blocks, XML-serialized <tool_use> calls, PII redaction, and AGPL-3.0 licensing.

Kimi K3

Provides an open-weight native multimodal agent that understands text and images within a 1,048,576-token context window for long-horizon coding, visual reasoning, and tool-driven workflows. Uses a 2.8T-parameter Mixture-of-Experts architecture (KDA + AttnRes) with MXFP4 quantization; best suited for research and large-scale inference setups.

Glint-Research/Fable-5-traces

A collection of 953 JSON-formatted Fable 5 interaction traces (includes chain-of-thought entries), published on Hugging Face under AGPL-3.0 — meant for fine-tuning or analyzing LLM behavior but subject to license and provenance constraints.

Qwable-v1

Open-weights agentic coding model that layers Claude Fable‑5 tool‑use SFT onto a reasoning‑distilled Qwen3.6 base; emits <tool_use> XML for file edits, shell commands and reads when prompted as an agent. Designed for agentic coding workflows; AGPL‑3.0 licensed.

microsoft/FastContext-1.0-4B-SFT

Provides a lightweight repository-exploration subagent for LLM coding agents: invoked on demand to run parallel read-only READ/GLOB/GREP calls and return compact file-path plus line-range citations so the main solver gets focused evidence instead of noisy reads.

Frames AI research as a trainable practice of reading, building, debugging, and fast feedback. The essay is most useful for researchers learning how to avoid hype-chasing, benchmark tunnel vision, and agent-induced blind spots.

Qwable 27B

Fine-tuned full checkpoint of the Qwen3.6-27B base that produces structured, trace-style assistant outputs for code, technical reasoning, and instruction-following. Packaged for local GGUF conversion and local inference; not a LoRA adapter and not validated for production use.

GLM-5.2

Provides a large language model optimized for long-horizon agentic tasks and end-to-end coding workflows — with a stable 1,000,000-token context, IndexShare sparse-attention and multi-level thinking-effort modes. MIT-licensed and designed for deployments that need sustained long-context reasoning and coding.

LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling

Uses Parallel Looped Transformers (PLT) to make loop count a practical knob for code models, finding two loops give the best test-time gains. Trains 7B models on 18T tokens and attributes saturation beyond two loops to a gain–cost tradeoff from positional mismatch.

GLM-5.2-FP8

Provides FP8-quantized weights of GLM-5.2 — a 744B long-context LLM tuned for sustained 1M-token engineering, coding and agentic workflows; compatible with vLLM, Transformers, SGLang and Ascend NPU deployments.

Guava: An Effective and Universal Harness for Embodied Manipulation

Provides a harness that lets language models control embodied manipulation via iterative perception–reasoning–action loops, semantic action abstractions, and multimodal observations. Demonstrates distilling capabilities into a 4B open-source model with under 2K simulated trajectories and shows sim-to-real generalization.

Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients

Proposes ZPPO, a distillation method that keeps the teacher inside prompts rather than injecting teacher gradients, using binary- and negative-candidate prompts plus a prompt replay buffer to recover learning signal on hard examples; shows gains for small Qwen3.5 students across 31 multimodal benchmarks.