Tag

Qwopus3.6-27B-v2-MTP

Fine-tuned reasoning model that speeds up structured multi-step outputs using Multi-Token Prediction (MTP) from a Qwen3.6-27B base. Produces more concise, faster generations for coding, DevOps, math, and constrained-format tasks; experimental community release for research and evaluation.

Miso TTS 8B

Generates conversational speech and voice continuation from text and optional audio context, outputting Mimi audio codes. Built on a Sesame-style CSM with an 8B Llama-like backbone plus a smaller autoregressive audio decoder. Suited for local TTS inference and voice-cloning workflows.

MiniCPM5-1B

A 1.08B-parameter causal LLM engineered for on-device text generation with native long-context (131k tokens) and built-in Think/No-Think modes. It emphasizes tool-calling support, lightweight deployment formats (BF16, GGUF, MLX), and RL+OPD post-training for stronger reasoning and code generation.

google/gemma-4-12B-it

Instruction-tuned, unified Gemma 4 12B multimodal model that accepts text, image and audio inputs and generates text outputs locally. Encoder-free design reduces multimodal latency and fits on consumer devices while offering long-context support and native thinking/system-prompt features.

Qwen3.7 Max Pi Traces

Provides raw newline-delimited JSON agent traces where assistant responses were generated by qwen/qwen3.7-max, captured with Teich; includes 47 JSONL files, an embedded tools schema snapshot, and conversion guidance for supervised fine‑tuning and distillation.

Gemma 4 12B Unified

A 12B unified, encoder-free multimodal model that directly ingests text, images and audio and returns text; supports very long contexts (up to 256K tokens), native function-calling/thinking modes, and small-model deployment for local or on-device use.

Step 3.7 Flash

Processes images and text to produce structured, reasoning-rich text outputs for high-throughput agentic workflows. Sparse MoE design (198B total, ~11B active per token), 256k context window and selectable reasoning levels—optimized for single-pass parsing, verification, and multi-step automation.

ADHD — a skill for coding agents

Spawns parallel, isolated LLM reasoning frames, then scores, clusters and prunes ideas to avoid premature convergence. Packaged as a reusable Claude/Codex agent skill with CLI and TypeScript APIs for ideation, design decisions and fuzzy debugging.

Mellum2 Thinking

Generates text with explicit chain-of-thought traces for multi-step reasoning and math-heavy tasks, emitting reasoning inside <think>...</think> blocks. Uses a Mixture-of-Experts design and 131k token context for long, verifiable workflows—best when you need inspectable reasoning.

nvidia/Qwen3.6-35B-A3B-NVFP4

Quantized NVFP4 build of the Qwen3.6-35B MoE language model, optimized with NVIDIA Model Optimizer to cut model size and GPU memory by ~3.06× for inference. Designed for vLLM and NVIDIA GPU deployments (Hopper/Blackwell).

A Matter of TASTE: Improving Coverage and Difficulty of Agent Benchmarks

Proposes TASTE, an automatic pipeline that synthesizes challenging agent benchmark tasks by sampling and evolving valid tool-sequence patterns; uses an adaptive contrastive n-gram model and LLM validity judgments to produce τ^c-Bench with broader tool-use coverage and higher difficulty.

SmoothConv

Provides ~100 hours of expert-annotated, multi-channel Chinese conversational speech with per-segment timestamps, speaker IDs and paralinguistic labels for turn-taking, overlap/interruption detection and full‑duplex dialogue research. Licensed for academic/non-commercial use (CC BY‑NC 4.0).