Agent skill

Quark Torch Export

Prepare export and downstream evaluation handoff for a planned or completed Quark PTQ run.

MITAuto-check passedAI & LLM Engineering

Install Quark Torch Export

skills CLI

$ npx skills add amd/Quark --skill quark-torch-export -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install amd/Quark quark-torch-export --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/amd/Quark.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills-impl/l1-atomic/torch/quark-torch-export .claude/skills/quark-torch-export && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "quark-torch-export" agent skill from https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export into .claude/skills/quark-torch-export/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quark-torch-export", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add amd/Quark --skill quark-torch-export -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install amd/Quark quark-torch-export --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/amd/Quark.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills-impl/l1-atomic/torch/quark-torch-export .agents/skills/quark-torch-export && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "quark-torch-export" agent skill from https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export into .agents/skills/quark-torch-export/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quark-torch-export", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add amd/Quark --skill quark-torch-export -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install amd/Quark quark-torch-export --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/amd/Quark.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills-impl/l1-atomic/torch/quark-torch-export .cursor/skills/quark-torch-export && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "quark-torch-export" agent skill from https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export into .cursor/skills/quark-torch-export/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quark-torch-export", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/amd/Quark.git --path .claude/skills-impl/l1-atomic/torch/quark-torch-export

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add amd/Quark --skill quark-torch-export -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install amd/Quark quark-torch-export --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/amd/Quark.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills-impl/l1-atomic/torch/quark-torch-export .gemini/skills/quark-torch-export && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "quark-torch-export" agent skill from https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export into .gemini/skills/quark-torch-export/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quark-torch-export", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install amd/Quark quark-torch-export

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add amd/Quark --skill quark-torch-export -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/amd/Quark.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills-impl/l1-atomic/torch/quark-torch-export .github/skills/quark-torch-export && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "quark-torch-export" agent skill from https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export into .github/skills/quark-torch-export/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quark-torch-export", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add amd/Quark --skill quark-torch-export -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install amd/Quark quark-torch-export --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/amd/Quark.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills-impl/l1-atomic/torch/quark-torch-export .opencode/skills/quark-torch-export && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "quark-torch-export" agent skill from https://github.com/amd/Quark/tree/release%2F0.13/.claude/skills-impl/l1-atomic/torch/quark-torch-export into .opencode/skills/quark-torch-export/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "quark-torch-export", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: quark-torch-export
GitHub stars: 181
Token cost: ~1.5k tokens
SKILL.md length: 458 words
Files: 1
Skills in repo: 37
Repo updated: Sep 28, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Prepare export and downstream evaluation handoff for a planned or completed Quark PTQ run.

Works in 3 steps: HuggingFace SafeTensors (hf_format) —… → ONNX (onnx) → GGUF (gguf)
The user wants to export a quantized model to HuggingFace SafeTensors
SKILL.md covers Purpose, Inputs, Outputs: run_manifest.yaml and Supported Export Formats, plus 6 more sections
Calls python

What it does

Quark Torch Export is an agent skill from amd/Quark. Prepare export and downstream evaluation handoff for a planned or completed Quark PTQ run. Use when the user wants to export a quantized model to HuggingFace SafeTensors, ONNX, or GGUF format, package for deployment, or set up post-quantization evaluation. Trigger for "export model", "save quantized model", "convert to GGUF", "export quantized model", "export to HuggingFace format", or when the user has a completed or planned PTQ run and needs deployment outputs.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering, covering LLM inference and serving, Model hubs and datasets and Deployment. It works with llama.cpp, ONNX and Hugging Face. The licence is MIT.

When your agent uses it

The user wants to export a quantized model to HuggingFace SafeTensors
Package for deployment
Set up post-quantization evaluation
Save quantized model

Example prompts

“export model”
“save quantized model”
“convert to GGUF”
“/quark-torch-export”

Requirements

Python 3

Workflow steps

3 steps, taken from the step headings in SKILL.md.

HuggingFace SafeTensors (hf_format) — Default
ONNX (onnx)
GGUF (gguf)

What it can do on your machine

Read from SKILL.md and the folder at commit 313cb0b. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
- python
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Quark Torch Export loads about 1.5k tokens when it runs. Until then it costs about 122 tokens; SKILL.md has 458 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~122
When it runs · the whole SKILL.md, loaded when a task matches: ~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from amd/Quark at commit 313cb0b, republished under its MIT licence (© amd). 458 words, ~1,464 tokens.

Download SKILL.mdSave it as .claude/skills/quark-torch-export/SKILL.md (or your agent's skills folder).

name: quark-torch-export
description: Prepare export and downstream evaluation handoff for a planned or completed Quark PTQ run. Use when the user wants to export a quantized model to HuggingFace SafeTensors, ONNX, or GGUF format, package for deployment, or set up post-quantization evaluation. Trigger for "export model", "save quantized model", "convert to GGUF", "export quantized model", "export to HuggingFace format", or when the user has a completed or planned PTQ run and needs deployment outputs.
layer: l1-atomic
primary_artifact: run_manifest.yaml
source_knowledge: examples/torch/language_modeling/llm_ptq/quantize_quark.py, quark/torch/export/api.py, examples/contrib/llm_eval/llm_eval.py

quark-torch-export

Purpose

Translate a confirmed quantization plan into export expectations and downstream evaluation requirements. Export is a post-quantization step — the model must be quantized first (or have a plan to be quantized) before export decisions make sense. This skill ensures the right export format is chosen and evaluation is properly configured.

Inputs

quant_plan.json from quark-torch-quant-plan
workspace_context.json for the output directory
run_manifest.yaml (optional, existing manifest to extend)

Outputs: run_manifest.yaml

Export does not own this artifact — it updates the workflow's manifest with export and evaluation config.

Schema: run_manifest.schema.json

(Export updates the workflow's run_manifest.yaml with export and evaluation fields rather than producing a new artifact.)

yaml

export:
  formats:
    - hf_format
  output_dir: ./output/qwen3-8b-fp8
  weight_format: real_quantized
  custom_mode: quark

evaluation:
  skip: false
  metrics:
    - ppl
  dataset: wikitext
  tasks: null
  batch_size: auto

Supported Export Formats

1. HuggingFace SafeTensors (`hf_format`) — Default

Produces: config.json + *.safetensors files with quantization_config metadata
Compatible with: HuggingFace transformers loading, vLLM, TGI
CLI flag: --model_export hf_format
Weight format options:
- real_quantized (default) — compressed, actual quantized weights
- fake_quantized — full-precision weights with quantization metadata only

2. ONNX (`onnx`)

Produces: quark_model.onnx with optimization passes applied
Compatible with: ONNX Runtime, TensorRT (with conversion)
CLI flag: --model_export onnx
Supports INT4/UINT4 conversion pass automatically

3. GGUF (`gguf`)

Produces: GGUF format file for llama.cpp and compatible inference engines
Requires: gguf>=0.10.0 package and tokenizer path
CLI flag: --model_export gguf
Best with: uint4_wo_32 scheme + AWQ algorithm

Multiple formats can be exported simultaneously: --model_export hf_format --model_export gguf

Export CLI Arguments

bash

python quantize_quark.py \
  --model_dir /path/to/model \
  --output_dir /path/to/output \
  --quant_scheme fp8 \
  --model_export hf_format \                    # Export format(s)
  --export_weight_format real_quantized \        # Compression mode
  --custom_mode quark \                          # Export mode: quark|awq|fp8
  --pack_method reorder                          # Weight packing: order|reorder

Evaluation Options

Post-quantization evaluation can be configured as part of the export step:

Perplexity (PPL)

Default dataset: wikitext
Flag: included by default unless --skip_evaluation is set

Task-Based Evaluation (via lm-eval harness)

bash

--tasks hellaswag,winogrande,arc_easy
--eval_batch_size auto
--num_fewshot 0

ROUGE/METEOR (for generation models)

bash

# Evaluated on cnn_dailymail by default
--use_mlperf_rouge  # For MLPerf-compatible ROUGE scoring

KV Cache Evaluation

bash

--use_ppl_eval_for_kv_cache
--ppl_eval_for_kv_cache_context_size 1024
--ppl_eval_for_kv_cache_sample_size 512

Skip Evaluation

bash

--skip_evaluation  # Skip all post-quantization evaluation

Model Reload for Separate Evaluation

If quantization and evaluation are done in separate steps:

bash

# Step 1: Quantize and export
python quantize_quark.py --model_dir MODEL --quant_scheme fp8 \
  --model_export hf_format --output_dir output/ --skip_evaluation

# Step 2: Reload and evaluate
python quantize_quark.py --model_dir MODEL --model_reload \
  --output_dir output/ --skip_quantization

Show full SKILL.md (193 more words)Show less

Rules

Never invent export paths that conflict with the existing run_manifest.yaml or quant_plan.json.
Export depends on quantization. If the model has not been quantized yet, this skill produces an export plan attached to the manifest — it does not run quantization.
Match export format to deployment target. Ask the user where the model will run: HuggingFace ecosystem → hf_format, llama.cpp → gguf, ONNX Runtime → onnx.
GGUF works best with UINT4. If the user wants GGUF but the plan uses FP8, flag the mismatch — GGUF is primarily designed for integer quantization.

Interaction Flow

Check prerequisites: Is there a quant_plan.json? Has quantization been run or is this plan-only?
Choose format: Ask where the model will be deployed and recommend the right export format.
Configure evaluation: Ask if the user wants post-quantization evaluation and which metrics.
Emit: Update run_manifest.yaml with export and evaluation configuration.

Recovery

If export is requested before quantization, return the missing prerequisites and attach the export request to the manifest under pending_exports.
If GGUF export fails, check that gguf>=0.10.0 is installed and that the tokenizer is accessible.
If ONNX export fails on a complex model, suggest trying hf_format first as a fallback.

Files

Just SKILL.md in .claude/skills-impl/l1-atomic/torch/quark-torch-export of amd/Quark.

Open the folder on GitHubat commit 313cb0b

Compare with similar skills

Quark Torch Export next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Quark Torch Export compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Quark Torch Export this skillamd/Quark	181	—	~1.5k	Automated safety check: Pass	MIT	9 days ago
SageMaker Serving Image Selectionhuggingface/skills	11k	1 repos	~4.6k	Automated safety check: Pass	Apache-2.0	6 days ago
Qwen Mtp GgufR6410418/Jackrong-llm-finetuning-guide	1.7k	—	~1.7k	Automated safety check: Pass	MIT	2 mo ago
Configure G1 Sim2realEGalahad/sim2real	145	—	~1.5k	Automated safety check: Pass	None	9 days ago
Add Modelguoqingbao/xinfer	333	—	~4.2k	Automated safety check: Notes	MIT	28 days ago
Hugging Face Local Modelshuggingface/skills	11k	3 repos	~945	Automated safety check: Pass	Apache-2.0	6 days ago

Similar skills

SageMaker Serving Image Selection
huggingface/skills
Official
Chooses the right serving container and current image URI for deploying a Hugging Face model to a SageMaker endpoint, preferring Hugging Face images over generic ones.
11k GitHub starsUsed in 1 repo~4.6k tokens
AI & LLM EngineeringAuto-check passed
Qwen Mtp Gguf
R6410418/Jackrong-llm-finetuning-guide
Complete agent-ready workflow for Qwen-family MTP or nextn GGUF conversion and release.
1.7k GitHub stars~1.7k tokensUpdated 2 mo ago
AI & LLM EngineeringAuto-check passed
Configure G1 Sim2real
EGalahad/sim2real
Install, repair, and verify sim2real on G1 robot computers. An agent skill from EGalahad/sim2real.
145 GitHub stars~1.5k tokensUpdated 9 days ago
AI & LLM EngineeringAuto-check passed
Add Model
guoqingbao/xinfer
Adapt and port new LLM model architectures to this xinfer project.
333 GitHub stars~4.2k tokensUpdated 28 days ago
AI & LLM EngineeringAuto-check: notes
Hugging Face Local Models
huggingface/skills
Official
Finds llama.cpp-compatible GGUF models on the Hugging Face Hub, picks a quantization for your hardware and launches them with llama-cli or llama-server.
11k GitHub starsUsed in 3 repos~945 tokens
AI & LLM EngineeringAuto-check passed
Resolve
alexziskind1/model-shelf
Always resolve Hugging Face models via model-shelf before any download.
130 GitHub stars~792 tokensUpdated 1 mo ago
AI & LLM EngineeringAuto-check passed

Questions about Quark Torch Export

What does Quark Torch Export do?

Prepare export and downstream evaluation handoff for a planned or completed Quark PTQ run. Quark Torch Export is an agent skill from amd/Quark. Prepare export and downstream evaluation handoff for a planned or completed Quark PTQ run.

When should I use Quark Torch Export?

Quark Torch Export fits situations like: the user wants to export a quantized model to HuggingFace SafeTensors; package for deployment; set up post-quantization evaluation; save quantized model.

How do I install Quark Torch Export in Claude Code?

Run `npx skills add amd/Quark --skill quark-torch-export -a claude-code`. Or copy the skill folder (.claude/skills-impl/l1-atomic/torch/quark-torch-export in amd/Quark) into .claude/skills/quark-torch-export in your project. Claude Code loads it when a task matches its description.

How do I install Quark Torch Export in Codex?

Run `npx skills add amd/Quark --skill quark-torch-export -a codex`. Or copy the skill folder (.claude/skills-impl/l1-atomic/torch/quark-torch-export in amd/Quark) into .agents/skills/quark-torch-export in your project. Codex loads it when a task matches its description.

Can I use Quark Torch Export in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add amd/Quark --skill quark-torch-export -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/quark-torch-export, .gemini/skills/quark-torch-export, .github/skills/quark-torch-export and .opencode/skills/quark-torch-export in your project.

What does Quark Torch Export need to run?

Going by SKILL.md and its folder, Quark Torch Export needs the command-line tools its instructions call (python). Our summary lists: Python 3.

Does Quark Torch Export access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Quark Torch Export safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Quark Torch Export use?

Quark Torch Export is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Quark Torch Export use?

About 1.5k tokens (SKILL.md is roughly 5.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Quark Torch Export?

Skills that share tags, products or a category with Quark Torch Export: SageMaker Serving Image Selection (huggingface/skills, 11k stars), Qwen Mtp Gguf (R6410418/Jackrong-llm-finetuning-guide, 1.7k stars), Configure G1 Sim2real (EGalahad/sim2real, 145 stars) and Add Model (guoqingbao/xinfer, 333 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Quark Torch Export?

amd (a GitHub organization) maintains it in amd/Quark, which has 181 GitHub stars. The repository holds 37 skills in this directory. The repository was last updated on September 28, 2026.

Source: amd/Quark on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Quark Torch Export

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

quark-torch-export

Purpose

Inputs

Outputs: run_manifest.yaml

Supported Export Formats

1. HuggingFace SafeTensors (hf_format) — Default

2. ONNX (onnx)

3. GGUF (gguf)

Export CLI Arguments

Evaluation Options

Perplexity (PPL)

Task-Based Evaluation (via lm-eval harness)

ROUGE/METEOR (for generation models)

KV Cache Evaluation

Skip Evaluation

Model Reload for Separate Evaluation

Rules

Interaction Flow

Recovery

Files

Compare with similar skills

Similar skills

More from amd/Quark

Related

Questions about Quark Torch Export

1. HuggingFace SafeTensors (`hf_format`) — Default

2. ONNX (`onnx`)

3. GGUF (`gguf`)