← All skills
ais-bench
80 / 100 · Low risk
Fine to try. Some of what it can do is expected: a coding skill usually runs build and test commands.
By ascend-ai-coding/awesome-ascend-skills (GitHub) · coding · 184 stars · scanned at commit 75ec89c
AISBench Benchmark - AI model evaluation tool for Ascend NPU. Supports accuracy evaluation (service/local models on text, multimodal datasets), performance evaluation (latency, throughput, stress testing, steady-state, real traffic simulation), vLLM/Triton inference services, 15+ benchmarks (MMLU, GSM8K, MMMU, docvqa, ocrbench_v2, etc.), multi-turn dialogue, Function Call (BFCL), and custom datasets.
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
The author does not mention this. Expected for a coding skill.
Found in 18 places.
Show 15 more
- skills/inference/ais-bench/SKILL.md, line 79
pip3 install -r requirements/api.txt - skills/inference/ais-bench/SKILL.md, line 96
ais_bench --models <model_task> --datasets <dataset_task> [--summarizer example] - skills/inference/ais-bench/SKILL.md, line 107
ais_bench --models vllm_api_general_chat --datasets demo_gsm8k_gen_4_shot_cot_chat_prompt --search - skills/inference/ais-bench/SKILL.md, line 149
ais_bench --models vllm_api_general_chat --datasets demo_gsm8k_gen_4_shot_cot_chat_prompt - skills/inference/ais-bench/SKILL.md, line 195
ais_bench --models vllm_api_general_chat --datasets custom_performance \ - skills/inference/ais-bench/SKILL.md, line 203
ais_bench --models vllm_api_general_chat --datasets sharegpt \ - skills/inference/ais-bench/SKILL.md, line 210
ais_bench --models vllm_api_general_chat --datasets custom \ - skills/inference/ais-bench/SKILL.md, line 221
ais_bench --models model1 model2 model3 --datasets dataset1 - skills/inference/ais-bench/SKILL.md, line 227
ais_bench --models model1 --datasets dataset1 dataset2 dataset3 - skills/inference/ais-bench/SKILL.md, line 233
ais_bench --models model1 model2 --datasets dataset1 dataset2 --parallel 4 - skills/inference/ais-bench/SKILL.md, line 306
ais_bench --models model1 --datasets dataset1 --resume outputs/default/20250628_151326 - skills/inference/ais-bench/SKILL.md, line 312
ais_bench --models model1 --datasets dataset1 --failed-only --resume outputs/default/20250628_151326 - skills/inference/ais-bench/SKILL.md, line 319
ais_bench --models model1 --datasets mmlu_merged - skills/inference/ais-bench/SKILL.md, line 325
ais_bench --models model1 --datasets dataset1 --repeat-n 5 - skills/inference/ais-bench/SKILL.md, line 358
bash scripts/check_env.sh
Downloads and installs other software
The author does not mention this.
Found in 6 places.
Show 3 more
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'ais-bench' ] || [ -L ~/.agents/skills/'ais-bench' ]; then echo 'A skill folder named ais-bench already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/ascend-ai-coding/awesome-ascend-skills' "$d" && git -C "$d" checkout --quiet '75ec89ce9d401be8ac95980e1c3f42f8d4315057' -- 'skills/inference/ais-bench' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/inference/ais-bench' ~/.agents/skills/'ais-bench' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'ais-bench' ~/.claude/skills/'ais-bench' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for ascend-ai-coding/awesome-ascend-skills shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/ascend-ai-coding/awesome-ascend-skills/)
Score history
| Date | Version | Score |
|---|
| 2026-11-08 | 75ec89c | 100 |