kali20gakki/msAgent / skills/quantizer/msmodelslim-adapter-verification
msmodelslim-adapter-verification
Verified为 msModelSlim 适配器执行功能性验证。适用于基础适配器开发完成后,自动执行四步验证(测试模型、全回退量化、权重一致性与可加载/保存、实际量化规则校验)并输出通过/失败结论。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.6 KB |
| references/dit/README.md | markdown | 3.2 KB |
| references/dit/fallback_config.yaml | config | 1.5 KB |
| references/dit/index.example.jsonl | other | 1.4 KB |
| references/dit/w8a8_dynamic_full_model.yaml | config | 1.3 KB |
| references/dit/w8a8_mxfp8_full_model.yaml | config | 1.6 KB |
| references/llm/fallback_config.yaml | config | 473 B |
| references/llm/w8a8_dynamic_full_model.yaml | config | 497 B |
| references/llm/w8a8_static_full_model.yaml | config | 497 B |
| references/verification_guide.md | markdown | 8.5 KB |
| references/vlm/fallback_config.yaml | config | 562 B |
| references/vlm/w8a8_dynamic_full_model.yaml | config | 652 B |
| references/vlm/w8a8_static_full_model.yaml | config | 652 B |
| scripts/step1_generate_test_model.py | script | 9.2 KB |
| scripts/step2_run_quantization.py | script | 3.4 KB |
| scripts/step2_run_quantization_dit.py | script | 14.4 KB |
| scripts/step3_verify_weights.py | script | 3.1 KB |
| scripts/step4_verify_quant_description.py | script | 5.0 KB |
references/dit/w8a8_mxfp8_full_model.yaml
View on GitHub# DiT 全模型 W8A8 MXFP8 量化模板(apiversion: multimodal_sd_modelslim_v1)
#
# step4 用:mxfp8 格式全模型量化。Wan2.2-T2V-A14B 实测配置(MindIE-SD × Ascend_950/Atlas_350)。
# 与 w8a8_dynamic_full_model.yaml 关键差异:
# - dtype: mxfp8(per_block 浮点),act/weight scope 均为 per_block
# - weight method: mse_round(act 仍用 minmax)
# 三条硬约束、inference_config 填写详见同目录 README.md。
# 占位符 _CALIB_DATASET_ 需替换;inference_config 按目标模型/任务调整字段。
apiversion: multimodal_sd_modelslim_v1
spec:
dataset: "_CALIB_DATASET_"
process:
- type: "linear_quant"
qconfig:
act:
scope: "per_block" # mxfp8 → per_block(int8 才用 per_token)
dtype: "mxfp8"
symmetric: True
method: "minmax"
weight:
scope: "per_block" # mxfp8 → per_block(int8 才用 per_channel)
dtype: "mxfp8"
symmetric: True
method: "mse_round"
include: ["*"] # 全模型量化
exclude: [] # 不回退
save:
- type: "mindie_format_saver"
part_file_size: 0
multimodal_sd_config:
dump_config:
enable_dump: False # MXFP8 是 data-free,无需 dump 校准统计量;保持显式声明
capture_mode: "args"
dump_data_dir: ""
inference_config: # 必须按目标推理仓补齐;以下为 Wan2.2-T2V-A14B 示例
size: "1280*720"
frame_num: 81
sample_steps: 40
convert_model_dtype: True
task: "t2v-A14B"Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |