asiai logo

{{ tr("hero_title") }}{{ tr("agent_hero_title") }}

{{ tr("hero_tagline") }}{{ tr("agent_hero_tagline") }}

{{ tr("mode_human") }} {{ tr("mode_agent") }}
{{ tr("cta_start") }} {{ tr("cta_leaderboard") }} {{ tr("cta_github") }} {{ tr("agent_cta_guide") }} {{ tr("agent_cta_api") }} {{ tr("agent_cta_copy") }}
Python 3.11+ Apache 2.0 Apple Silicon {{ tr("agent_badge") }}
asiai CLI benchmark

asiai bench

asiai web dashboard

asiai web

GET /api/status ≤ 500ms 
{
  "chip": "Apple M4 Pro",
  "ram_gb": 64.0,
  "memory_pressure": "normal",
  "gpu_utilization_percent": 45.2,
  "engines": {
    "ollama": { "running": true, "models_loaded": 2 },
    "lmstudio": { "running": true, "models_loaded": 1 }
  }
}
GET /api/snapshot Full state 
{
  "system": {
    "chip": "Apple M4 Pro",
    "gpu_cores": 20,
    "gpu_utilization_percent": 45.2,
    "gpu_renderer_percent": 38.1,
    "thermal_state": "nominal"
  },
  "engines": [{
    "name": "ollama",
    "models": [{ "name": "qwen3.5:latest", "size_params": "35B" }]
  }]
}

{{ tr("pain_title") }}

{{ tr("pain_subtitle") }}

{{ tr("pain1_title") }}

{{ tr("pain1_desc") }}

{{ tr("pain2_title") }}

{{ tr("pain2_desc") }}

{{ tr("pain3_title") }}

{{ tr("pain3_desc") }}

{{ tr("feat_title") }}

{{ tr("feat_subtitle") }}

{{ tr("feat1_t") }}

{{ tr("feat1_d") }}

{{ tr("feat3_t") }}

{{ tr("feat3_d") }}

{{ tr("feat5_t") }}

{{ tr("feat5_d") }}

{{ tr("feat10_t") }}

{{ tr("feat10_d") }}

{{ tr("feat14_t") }}

{{ tr("feat14_d") }}

{{ tr("feat13_t") }}

{{ tr("feat13_d") }}

{{ tr("uc_title") }}

{{ tr("uc_subtitle") }}

"{{ tr("uc1_t") }}"

{{ tr("uc1_d") }}

"{{ tr("uc2_t") }}"

{{ tr("uc2_d") }}

"{{ tr("uc3_t") }}"

{{ tr("uc3_d") }}

"{{ tr("uc4_t") }}"

{{ tr("uc4_d") }}

"{{ tr("uc5_t") }}"

{{ tr("uc5_d") }}

"{{ tr("uc6_t") }}"

{{ tr("uc6_d") }}

"{{ tr("uc7_t") }}"

{{ tr("uc7_d") }}

"{{ tr("uc8_t") }}"

{{ tr("uc8_d") }}

"{{ tr("uc9_t") }}"

{{ tr("uc9_d") }}

"{{ tr("uc10_t") }}"

{{ tr("uc10_d") }}

"{{ tr("uc11_t") }}"

{{ tr("uc11_d") }}

"{{ tr("uc12_t") }}"

{{ tr("uc12_d") }}

{{ tr("how_title") }}

{{ tr("how_subtitle") }}

1

{{ tr("step1") }}

brew install asiai
2

{{ tr("step2") }}

$ asiai detect ✔ ollama (11434) ✔ lmstudio (1234) ✔ mlx-lm (8080) → {{ tr("step2_out") }}
3

{{ tr("step3") }}

$ asiai bench -m qwen3.5 Engine tok/s TTFT lmstudio 71.2 42ms ollama 54.8 61ms mlx-lm 30.1 38ms

{{ tr("disc_title") }}

{{ tr("disc_subtitle") }}

2.3x

MLX vs llama.cpp

{{ tr("disc1_d") }}

Flat

{{ tr("disc2_t") }}

{{ tr("disc2_d") }}

30 vs 71

{{ tr("disc3_t") }}

{{ tr("disc3_d") }}

{{ tr("eng_title") }}

{{ tr("eng_subtitle") }}

Engine {{ tr("eng_port") }} API {{ tr("eng_format") }} VRAM
Ollama 11434 {{ tr("eng_native") }} GGUF ✔
LM Studio 1234 {{ tr("eng_compat") }} GGUF + MLX ✔
mlx-lm 8080 {{ tr("eng_compat") }} MLX —
llama.cpp 8080 {{ tr("eng_compat") }} GGUF —
oMLX 8000 {{ tr("eng_compat") }} MLX —
vllm-mlx 8000 {{ tr("eng_compat") }} MLX —
Exo 52415 {{ tr("eng_compat") }} MLX —

{{ tr("met_title") }}

{{ tr("met_subtitle") }}

tok/s

{{ tr("m_toks") }}

TTFT

{{ tr("m_ttft") }}

Power (W)

{{ tr("m_power") }}

tok/s/W

{{ tr("m_eff") }}

Stability

{{ tr("m_stab") }}

VRAM

{{ tr("m_vram") }}

Thermal

{{ tr("m_therm") }}

Context

{{ tr("m_ctx") }}

{{ tr("cta_title") }}

{{ tr("cta_subtitle") }}

Homebrew
brew tap druide67/tap brew install asiai
pip
pip install asiai
GitHub {{ tr("cta_doc") }} {{ tr("cta_method") }} Apache 2.0 ❤ Sponsor

⭐ {{ tr("cta_star") }}

{% if lang in ("zh", "ja", "ko") %}
{% if lang == "zh" %} 🌍 本页翻译由机器生成，可能不完全准确。欢迎通过 GitHub Issue 提交修正。 {% elif lang == "ja" %} 🌍 このページは機械翻訳です。正確でない場合があります。修正は GitHub Issue でお寄せください。 {% elif lang == "ko" %} 🌍 이 페이지는 기계 번역입니다. 정확하지 않을 수 있습니다. 수정 사항은 GitHub Issue로 제출해 주세요. {% endif %}
{% endif %}