What is hailo-optimize?

A model-conversion step that optimizes and quantizes a Hailo HAR file for a Hailo hardware accelerator. Quantization reduces the numerical precision used by a model so it can run on the target device.

It helps use calibration data, apply normalization and post-processing settings, and create an optimized HAR file for Hailo10H, Hailo15H, or Hailo15L hardware.

It prepares a parsed model for deployment by applying the matching model settings and checking the result against a signal-quality threshold.

How hailo-ai/hailo_model_zoo configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to hailo-ai/hailo_model_zoo. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/hailo-ai/hailo_model_zoo/master/.claude/skills/hailo-optimize/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/hailo-ai/hailo_model_zoo

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/hailo-ai/hailo_model_zoo/hailo-optimize/github.svg)](https://agentmods.dev/skills/hailo-ai/hailo_model_zoo/hailo-optimize)

Your own site

<a href="https://agentmods.dev/skills/hailo-ai/hailo_model_zoo/hailo-optimize"><img src="https://agentmods.dev/badge/skills/hailo-ai/hailo_model_zoo/hailo-optimize/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/hailo-ai/hailo_model_zoo/hailo-optimize"><img src="https://agentmods.dev/badge/skills/hailo-ai/hailo_model_zoo/hailo-optimize.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 67 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,410 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00067	$0.03410
Opus 5.5	$0.00027	$0.01364
Sonnet 5.5	$0.00013	$0.00682
Haiku 4.5	$0.00007	$0.00341

Measured 5d ago against content hash 587a5dbebe5e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

hailo-optimize scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/hailo-optimize/SKILL.md · 200 lines

How it starts

The opening of the file, as written. The whole thing — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hailo Optimize

Second stage of the DFC flow: take a parsed .har from /hailo-parse and produce a quantized <model_name>_optimized.har. This step applies a model script (ALLS commands — normalization, optimization flavor, NMS postprocess, fine-tune, etc.) and runs the DFC's quantization with a calibration set.

This skill is shipped with the Hailo Model Zoo. It assumes the DFC wheel (hailo_sdk_client) is installed in the active virtualenv.

Inputs

Arg	Required	Notes
`<parsed.har>`	yes	Output of `/hailo-parse`.
`[hw_arch]`	no	One of `hailo10h` (default), `hailo15h`, `hailo15l`.

Follow-up questions: calibration source (numpy .npy / .npz, image directory, or list of image paths), and whether to use a custom ALLS instead of the auto-matched one.

Workflow

1. Resolve inputs

Verify <parsed.har> exists.
model_name = HAR filename stem (strip a trailing _parsed if present).
hw_arch defaults to hailo10h.
Confirm hailo_sdk_client import works.

2. Auto-match ALLS (silent)

Look up an ALLS file in this priority order:

hailo_model_zoo/cfg/alls/<hw_arch>/performance/<model_name>.alls
hailo_model_zoo/cfg/alls/<hw_arch>/base/<model_name>.alls
hailo_model_zoo/cfg/alls/generic/<model_name>.alls

hailo10h has no dedicated subdir — for it, only generic/ is consulted.

Apply the same name-matching rules as /hailo-parse (case-insensitive, -↔_, version-suffix stripping).

On a hit: load the file as the model script and log Using Model Zoo ALLS: hailo_model_zoo/cfg/alls/<...>/<name>.alls.

On a miss: compose a minimal default and show it to the user before applying:

normalization1 = normalization([0.0, 0.0, 0.0], [255.0, 255.0, 255.0])
model_optimization_flavor(optimization_level=2)

(Use the mean_list / std_list from the matched parser YAML, if /hailo-parse passed them along; otherwise the 0/255 defaults assume image input in 0–255 range.)

3. Model-script command reference

Read the full file on GitHub · 200 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

5d ago Changed · +37 lines 587a5dbebe5e
2mo ago First seen · 163 lines · 67 tokens per session scan A 3074d47bef3d

Subscribe to this mod's changes

hailo-optimize is a skill published in the GitHub repository hailo-ai/hailo_model_zoo (722 stars, last pushed 5d ago), licensed MIT. It adds 67 tokens to every session and 3,410 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Data and AI Hardware and IoT ai-accelerators computer-vision deep-learning edge-ai hailo8 quantized-neural-networks

hailo-optimize next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
hailo-optimizehailo-ai/hailo_model_zoo	Skill	722	5d ago	MIT	A	67	Claude Code
yolo-exportultralytics/skills	Skill	35	yesterday	AGPL-3.0	A	120	Claude Code, Codex
voyager-launchfxd0h/Axelera-Voyager-Local-Assistant	Skill	4	3mo ago	MIT	A	162	Claude Code
voyager-list-modelsfxd0h/Axelera-Voyager-Local-Assistant	Skill	4	3mo ago	MIT	A	39	Claude Code
voyager-new-pipelinefxd0h/Axelera-Voyager-Local-Assistant	Skill	4	3mo ago	MIT	A	71	Claude Code
voyager-add-modelfxd0h/Axelera-Voyager-Local-Assistant	Skill	4	3mo ago	MIT	A	45	Claude Code
voyager-deployfxd0h/Axelera-Voyager-Local-Assistant	Skill	4	3mo ago	MIT	A	74	Claude Code

hailo_model_zoo: Skill for Claude Code