实践之Hermes

Posted on | In | Visitors
Hermes Agent Hermes Agent (noun search) Hermes中文社区 The AI agent that grows with you.

Install it on your server, give it your messaging accounts, and it becomes a persistent personal agent — learning your projects, building its own skills, and reaching you wherever you are. Not a chatbot. Not a copilot. An agent that lives on your machine and gets smarter every day.

Install

1
$ curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash

Installs uv, Python 3.11, clones the repo, sets up everything. No sudo needed.

Configure

# Interactive setup wizard

1
$ hermes setup

# Or choose your model

1
$ hermes model

Connect to Nous Portal (OAuth), OpenRouter (API key), or your own endpoint. Interactive setup wizard.

配置hermes，使用硅基流动的deepseek-ai/DeepSeek-V4-Pro，例如：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
$ hermes model

  Current model:    MiniMax-M2.7-highspeed
  Active provider:  MiniMax


Custom OpenAI-compatible endpoint configuration:

API base URL [e.g. https://api.example.com/v1]: https://api.siliconflow.cn/v1
API key [optional]: ***************************************************
                                                                       Verified endpoint via https://api.siliconflow.cn/v1/models (93 model(s) visible)

Select API compatibility mode:

    1. Auto-detect [current]
       Use Hermes URL heuristics; best for standard OpenAI-compatible endpoints.
    2. Chat Completions
       Use /chat/completions for standard OpenAI-compatible servers.
    3. Responses / Codex
       Use /responses for Codex-compatible tool-calling backends.
    4. Anthropic Messages
       Use /v1/messages for Anthropic-compatible endpoints.
       Choice [1-4, Enter to keep current/detected]: 1
         API mode: auto-detect
         Available models:
              1. deepseek-ai/DeepSeek-V4-Pro
                 2. deepseek-ai/DeepSeek-V4-Flash
                    3. Pro/moonshotai/Kimi-K2.6
                    4. Pro/zai-org/GLM-5.1
                    5. nex-agi/Nex-N2-Pro
                    6. MiniMaxAI/MiniMax-M2.5
                    7. Pro/MiniMaxAI/MiniMax-M2.5
                    8. Tongyi-MAI/Z-Image-Turbo
                    9. Tongyi-MAI/Z-Image
                    10. baidu/ERNIE-Image-Turbo
                    11. Pro/zai-org/GLM-5
                    12. Pro/moonshotai/Kimi-K2.5
                    13. Pro/zai-org/GLM-4.7
                    14. deepseek-ai/DeepSeek-V3.2
                    15. Pro/deepseek-ai/DeepSeek-V3.2
                    16. deepseek-ai/DeepSeek-V3.1-Terminus
                    17. Pro/deepseek-ai/DeepSeek-V3.1-Terminus
                    18. Qwen/Qwen3.6-35B-A3B
                    19. Qwen/Qwen3.6-27B
                    20. Qwen/Qwen3.5-397B-A17B
                    21. Qwen/Qwen3.5-122B-A10B
                    22. Qwen/Qwen3.5-35B-A3B
                    23. Qwen/Qwen3.5-27B
                    24. Qwen/Qwen3.5-9B
                    25. Qwen/Qwen3.5-4B
                    26. Qwen/Qwen3-VL-Embedding-8B
                    27. Qwen/Qwen3-VL-Reranker-8B
                    28. PaddlePaddle/PaddleOCR-VL-1.5
                    29. deepseek-ai/DeepSeek-R1
                    30. Pro/deepseek-ai/DeepSeek-R1
                    31. deepseek-ai/DeepSeek-V3
                    32. Pro/deepseek-ai/DeepSeek-V3
                    33. stepfun-ai/Step-3.5-Flash
                    34. Qwen/Qwen3-VL-32B-Instruct
                    35. Qwen/Qwen3-VL-32B-Thinking
                    36. Qwen/Qwen3-VL-8B-Instruct
                    37. Qwen/Qwen3-VL-8B-Thinking
                    38. Qwen/Qwen3-VL-30B-A3B-Instruct
                    39. Qwen/Qwen3-VL-30B-A3B-Thinking
                    40. Qwen/Qwen3-Omni-30B-A3B-Instruct
                    41. Qwen/Qwen3-Omni-30B-A3B-Thinking
                    42. Qwen/Qwen3-Omni-30B-A3B-Captioner
                    43. deepseek-ai/DeepSeek-OCR
                    44. inclusionAI/Ling-flash-2.0
                    45. inclusionAI/Ling-mini-2.0
                    46. Qwen/Qwen-Image-Edit-2509
                    47. Qwen/Qwen-Image-Edit
                    48. Qwen/Qwen-Image
                    49. tencent/Hunyuan-MT-7B
                    50. ByteDance-Seed/Seed-OSS-36B-Instruct
                    51. Wan-AI/Wan2.2-I2V-A14B
                    52. Wan-AI/Wan2.2-T2V-A14B
                    53. zai-org/GLM-4.5V
                    54. zai-org/GLM-4.5-Air
                    55. TeleAI/TeleSpeechASR
                    56. Qwen/Qwen3-Coder-30B-A3B-Instruct
                    57. Qwen/Qwen3-30B-A3B-Instruct-2507
                    58. tencent/Hunyuan-A13B-Instruct
                    59. deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
                    60. Qwen/Qwen3-32B
                    61. Qwen/Qwen3-14B
                    62. Qwen/Qwen3-8B
                    63. Qwen/Qwen3-Reranker-8B
                    64. Qwen/Qwen3-Embedding-8B
                    65. Qwen/Qwen3-Reranker-4B
                    66. Qwen/Qwen3-Embedding-4B
                    67. Qwen/Qwen3-Reranker-0.6B
                    68. Qwen/Qwen3-Embedding-0.6B
                    69. THUDM/GLM-4-32B-0414
                    70. THUDM/GLM-Z1-9B-0414
                    71. THUDM/GLM-4-9B-0414
                    72. fnlp/MOSS-TTSD-v0.5
                    73. FunAudioLLM/CosyVoice2-0.5B
                    74. FunAudioLLM/SenseVoiceSmall
                    75. BAAI/bge-m3
                    76. BAAI/bge-reranker-v2-m3
                    77. netease-youdao/bce-embedding-base_v1
                    78. netease-youdao/bce-reranker-base_v1
                    79. Kwai-Kolors/Kolors
                    80. Qwen/Qwen2.5-72B-Instruct-128K
                    81. Qwen/Qwen2.5-72B-Instruct
                    82. Qwen/Qwen2.5-32B-Instruct
                    83. Qwen/Qwen2.5-14B-Instruct
                    84. Qwen/Qwen2.5-7B-Instruct
                    85. BAAI/bge-large-en-v1.5
                    86. BAAI/bge-large-zh-v1.5
                    87. Pro/BAAI/bge-m3
                    88. Pro/Qwen/Qwen2.5-7B-Instruct
                    89. Pro/BAAI/bge-reranker-v2-m3
                    90. LoRA/Qwen/Qwen2.5-32B-Instruct
                    91. LoRA/Qwen/Qwen2.5-14B-Instruct
                    92. LoRA/Qwen/Qwen2.5-72B-Instruct
                    93. LoRA/Qwen/Qwen2.5-7B-Instruct
                          Select model [1-93] or type name: 1
                        Context length in tokens [leave blank for auto-detect]: 
                        Display name [Api.siliconflow.cn]: deepseek-ai/DeepSeek-V4-Pro[siliconflow]
                        Default model set to: deepseek-ai/DeepSeek-V4-Pro (via https://api.siliconflow.cn/v1)
                          💾 Saved to custom providers as "deepseek-ai/DeepSeek-V4-P[siliconflow]" (edit in config.yaml)

Start Chatting

1
$ hermes

Full interactive CLI with tools, memory, and skills. That’s it.

Go Multi-Platform (Optional)

Interactive gateway setup wizard

1
$ hermes gateway setup

Wechat

1
$ hermes pairing approve weixin MA8M8MR9

Start the messaging gateway

1
$ hermes gateway

Install as a system service

1
$ hermes gateway install

Walk through connecting Telegram, Discord, Slack, or WhatsApp. Runs as a systemd service.

Keep It Up to Date

1
$ hermes update

Pulls the latest changes and reinstalls dependencies. Run anytime to get new features and fixes.

dashboard

1
$ hermes dashboard