BugTraceAI-Apex-G4-26B-Q4
- 105
- 3.6k
- apache-2.0
- 6 mo ago
Feature extraction model, used with comfyui.
Gated access
A pure C inference engine designed for Intel Xe-LP iGPUs and Gemma 4 26B-A4B QAT.
Gemma 4 31B Abliterated — quality-preserving guardrail removal for Google's most capable open model. Apache 2.0. Runs on Apple Silicon via MLX.
Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode.
Run Google's Gemma 4 entirely in your browser via WebGPU. Multimodal chat, E2B vs E4B arena, ONNX conversion toolkit.
Xiaofeiyu (小肥鱼) — Gemma 4 Split-Phase 5-Slider Spoken Persona Engine: 100% Base Coding IQ + Real-Time Mandarin IM Bubbles on 12GB GPUs, Built-in 2D Persona Orb Web UI & DSH Plugin
A simplified functionality reproduction of Jev-style decisions with Gemma 4. Native multi-modal ability, finetuning free, runs locally on your machine.
Signed, auditable receipts for Gemma 4 inference. The provider commits to the execution trace; an auditor challenges random positions and verifies them on a CPU without the weights: Freivalds checks, exact decode, Merkle-bound traces. CommitLLM protocol.
Gemma 4 Indonesian recipe fine-tune case study
Runs in the browser on Hugging Face
Gemma 4 on a 24GB MacBook: measured recipes, runtimes, and fallback paths.
Distilling a small Gemma 4 guard model for AI agent tool calls: LoRA adapter, sanitized dataset, replay evaluation and honest results against Gemini guards
A multimodal support-to-engineering agent that investigates bug reports, reproduces issues, patches code, and verifies fixes. Winner of the Enterprise Impact track at the Cerebras × Google DeepMind hackathon.
Nemotron-3-Ultra-550B, Gemma4-31B on 8xA100 40GB GPUs as production AWS SageMaker endpoint with tensor-parallel inference, Gemma4-12B-it on Lambda GPU, Megatron-Bridge w vLLM or SGLang,
An on-device AI supermarket inventory and pantry management application built with Flutter, Google's LiteRT-LM (gemma-4-E2B-it.litertlm), Google ML Kit Barcode Scanning, and SQLite
A Flutter Edge AI application demonstrating on-device, zero-shot multimodal object segmentation and real-time tracking. Combines Gemma 4 w/ EdgeTAM & DINOv3
CITADEL — 13-layer open AI evaluation infrastructure. Gemma 4 27B benchmarked against 5 frontier models with Ed25519-signed audit chains and EU AI Act / NIST / ISO 42001 / HIPAA compliance reports. Gemma 4 Good Hackathon submission. MIT.
One-notebook Gemma 4 RL on Apple Silicon.