Accuracy Safe Quantization

Shrink a converted LiteRT model with ai-edge-quantizer (fp16 / int8 / int4) without losing accuracy, verifying parity against the float source after every step. Use when choosing a quantization recipe for a new model, when a quantized model fails to load, degrades on a task benchmark, or degenerates over long generations, or when deciding between dynamic-range, weight-only, and blockwise variants.

499 stars 134 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add google-ai-edge/litert-samples --skill accuracy-safe-quantization
Manual install
git clone --depth 1 https://github.com/google-ai-edge/litert-samples.git /tmp/litert-samples
mkdir -p ~/.claude/skills
cp -r /tmp/litert-samples/skills/accuracy-safe-quantization ~/.claude/skills/accuracy-safe-quantization

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Accuracy-safe quantization

A quantization is done when three things hold, in this order:

1. it exports and the file shrinks by what the recipe predicts,
2. **output parity with the float source holds on a task-level check**, not
   just a smoke test,
3. the quantized model still passes the deployment check on the target
   runtime and device.

Quantization rewrites the graph, so step 3 is a fresh obligation every time:
re-run the same CompiledModel verification you used to accept the float
conversion (see the `gpu-clean-conversion` skill), then the on-device
numerical check.

All recipes below are `ai-edge-quantizer` (`pip install ai-edge-quantizer`;
0.9.0 as of 2026-10, which pins `ai-edge-litert` 2.2.0 on the host),
plain Python, no build step. Worked examples live in this repo under
`models/bonsai/bonsai_image_4b/converted/` and
`models/qwen/qwen3_tts/converted/`.

## Choosing a lane

Start with the lightest recipe that meets the size budget, and move down
only on evidence:

| Budget / model | Recipe |
|---|---|
| ~2× smaller, zero risk | **fp16 float-casting.** Weights cast to fp16, compute stays float. On a GPU that already computes in fp16 this is close to free numerically — verify anyway |
| ~4× smaller — encoders, conv nets, diffusion blocks | **Dynamic-range int8 channelwise.** int8 weights, float activations; this shape rides the GPU delegate |
Read the full skill on GitHub

Details

Skill name
accuracy-safe-quantization
Path
skills/accuracy-safe-quantization/SKILL.md
License
Apache-2.0
Repo updated
17 November 2026

FAQ

What does the Accuracy Safe Quantization skill do?

Shrink a converted LiteRT model with ai-edge-quantizer (fp16 / int8 / int4) without losing accuracy, verifying parity against the float source after every step. Use when choosing a quantization recipe for a new model, when a quantized model fails to load, degrades on a task benchmark, or degenerates over long generations, or when deciding between dynamic-range, weight-only, and blockwise variants.

How do I install the Accuracy Safe Quantization skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add google-ai-edge/litert-samples --skill accuracy-safe-quantization. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Accuracy Safe Quantization popular?

It lives in google-ai-edge/litert-samples, which has 499 GitHub stars and 134 forks.

Related skills

Litert Lm

499

google-ai-edge/litert-samples

Builds an Android app that runs an open text LLM on the device, on the CPU or the GPU, with the LiteRT-LM Kotlin API (com.google.ai.edge.litertlm:litertlm-android). Use this skill when creating a chat, summarization, or extraction feature with a .litertlm model from Hugging Face litert-community (Gemma, Qwen, Llama, Phi), including the Gradle dependency, the GPU manifest entries, getting the model file onto the device, Engine initialization off the main thread, a streamed multi-turn Conversation with sendMessageAsync, and the AndroidViewModel and Compose screen. Don't use for running .tflite v

Mobile Development134

Litert Runtime

499

google-ai-edge/litert-samples

Configures and implements LiteRT (.tflite) on-device inference on Android in Kotlin using the CompiledModel API on CPU and GPU. Use this skill for any Android LiteRT or .tflite task, including Gradle setup, image classification or vision/audio/embedding helpers, TensorBuffer reuse, CPU memory optimization, Bitmap preprocessing, AndroidViewModel GPU-to-CPU fallback with warm-up on a single-thread dispatcher, LiteRT in Google Play services (play-services-litert), migrating from legacy org.tensorflow.lite.Interpreter, GpuDelegate, or play-services-tflite, and latency benchmarking. Don't use for r

Mobile Development134

Benchmark on Ddp

499

google-ai-edge/litert-samples

Measure a LiteRT .tflite model with `benchmark_model` on Developer Device Platform (DDP) lab phones through `litert benchmark --ddp`, or on the Mac you run it from, and turn the session into rows of the performance leaderboard under benchmark/leaderboard - one matrix entry per model, one session per accelerator, collect results.pb and runtime_info.pb, rebuild board.json, check the page, commit the data. Use when a model needs latency and memory numbers per platform, device and accelerator, or when a new benchmark_model release means re-measuring the board.

AI, LLMs & Agents134

AI Discoverability Audit

457

BrianRWagner/ai-marketing-claude-code-skills

Use when a founder, marketer, or consultant wants to audit how visible their brand is to AI search (ChatGPT, Perplexity, Claude, Gemini). Also use when mentioned: 'AI SEO', 'GEO', 'ChatGPT can't find me', 'AI search visibility', 'how do I show up in AI answers'. NOT for: traditional Google SEO, ad strategy, or content writing.

AI, LLMs & AgentsIn 1 list105

Compiled Model App Scaffolding

499

google-ai-edge/litert-samples

Build a new Android app (Kotlin, Compose) around a verified LiteRT model using the CompiledModel API - the app architecture, the inference-layer lifecycle rules, model delivery, and the UI traps that masquerade as model bugs. Use when turning a converted and device-verified model into a demo or product app, when an app's inference layer leaks memory or blocks the UI, or when a model that verified clean looks wrong inside an app.

AI, LLMs & Agents134

Workbuddy Wiki

757

chujianyun/skills

WorkBuddy(腾讯云 CodeBuddy 团队的 AI Agent 桌面办公工作台,俗称小龙虾)官方文档离线 Wiki,覆盖 workbuddy 桌面端、workbuddyapp 手机 App、workbuddymini 小程序与企业版概述。当用户询问 WorkBuddy 的安装、快速开始、Craft/Plan/Ask 模式、任务管理、Skills/MCP、连接器、知识库、定时任务、模型切换、计费积分、企业微信等渠道接入或故障排查时使用。企业版后台管理文档需登录,未纳入本快照。

AI, LLMs & Agents102

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.