Supertonic 3: What On-Device TTS Starts to Look Like When ONNX Is the Contract
A source audit of Supertonic 3 by Supertone AI, analyzing 99M-parameter ONNX Runtime TTS, cross-platform inference, and local OpenAI-compatible speech serving.
Fodev JEO
LATEST INVESTIGATIONS
Source audits & field notes
Unsloth Local API Guide: Run Claude/OpenAI-Style Clients on Your Own Machine
A practical, code-first guide to exposing local GGUF models as authenticated Anthropic/OpenAI-compatible APIs with Unsloth.
RLM-FORGE: Architecting Evidence-Gated Recursion in Multi-Agent Loops
A source audit of Q00/rlm-forge, deconstructing Ouroboros recursion orchestration, Hermes Agent sandboxing, and TraceGuard deterministic evidence contracts.
LLM Serving Repositories Target Different Layers
A pinned-source audit finds that fifteen repositories span serving engines, runtimes, frameworks, kernels, training, and speech recognition—not one comparable speed ranking.
CLI-Anything Architecture: Why Agent-Native Interfaces Outweigh Model Scaling
A source audit of HKUDS’s CLI-Anything, analyzing agent-native harness engineering, CLI-Hub package distribution, deterministic execution, and multimodal CAD pipelines.
SuperGemma4's Perfect Code Score Misses a Syntax Error
A pinned audit of SuperGemma4’s 20-prompt quick bench finds one 10/10 code answer Python cannot parse, while baseline gains remain unreproducible.
Sentence Transformers v6.1 Makes Input Order Part of Your Index
Sentence Transformers keeps familiar calls, but v6.1 changes structured-message ordering. A pinned audit separates index compatibility, Qwen prompts and reranker scores.
Feynman Research Recovery Is Not Evidence Assurance
A pinned source audit of Feynman v0.5.47 separates signed-in alpha tools, abstract-only fallbacks, prompt context, and conditional telemetry error capture.
AutoBE Release Diff: 133 of 134 Compiler Files Match
Source-pinned audit of AutoBE releases, write-before-complete guards, and realization-test cleanup boundaries—not a reliability benchmark.
LLM Wiki: Why Agent-Maintained Markdown Outperforms Ephemeral Vector RAG
An architectural evaluation of Andrej Karpathy’s LLM Wiki, comparing persistent markdown compilation against ephemeral RAG retrieval and analyzing 2026 wiki harnesses.
인기 태그
Search article titles, categories, and tags. Full text is searched when needed.
Type to search articles.