Bug Description
错误信息提示可以通过设置 model.context_length 来覆盖,但实际上这个设置被忽略了。
复现步骤
- 在 config.yaml 中设置:
model:
default: Qwen3-235B-A22B
provider: custom
base_url: https://api.scnet.cn/api/llm/v1
api_mode: chat_completions
context_length: 32768
- 启动 Hermes Agent
- 尝试发送消息,会收到上下文窗口大小错误
期望行为
当用户在 config.yaml 中明确设置了 model.context_length 时,应该尊重用户的设置,绕过最小长度检查。
实际行为
即使设置了 model.context_length,仍然会检查这个值是否小于 64,000 tokens,并触发错误。
解决方案
修改 run_agent.py 文件中的上下文长度检查逻辑,在用户明确设置了 context_length 时跳过最小长度检查:
# Skip minimum context length check if user explicitly set context_length in config
if _ctx and _ctx < MINIMUM_CONTEXT_LENGTH and self._config_context_length is None:
raise ValueError(
f"Model {self.model} has a context window of {_ctx:,} tokens, "
f"which is below the minimum {MINIMUM_CONTEXT_LENGTH:,} required "
f"by Hermes Agent. Choose a model with at least "
f"{MINIMUM_CONTEXT_LENGTH // 1000}K context, or set "
f"model.context_length in config.yaml to override."
)
环境信息
- Hermes Agent 版本:最新版本
- 操作系统:Linux
- 模型:Qwen3-235B-A22B (上下文窗口大小:32,768 tokens)
Steps to Reproduce
Model Qwen3-235B-A22B has a context window of 32,768 tokens, which is below the minimum 64,000 required by Hermes Agent. Choose a model with at least 64K context, or set model.context_length in config.yaml to override.
错误信息提示可以通过设置 `model.context_length` 来覆盖,但实际上这个设置被忽略了。
## 复现步骤
1. 在 config.yaml 中设置:
```yaml
model:
default: Qwen3-235B-A22B
provider: custom
base_url: https://api.scnet.cn/api/llm/v1
api_mode: chat_completions
context_length: 32768
- 启动 Hermes Agent
- 尝试发送消息,会收到上下文窗口大小错误
Expected Behavior
当用户在 config.yaml 中明确设置了 model.context_length 时,应该尊重用户的设置,绕过最小长度检查。
Actual Behavior
即使设置了 model.context_length ,仍然会检查这个值是否小于 64,000 tokens,并触发错误
Affected Component
Gateway (Telegram/Discord/Slack/WhatsApp)
Messaging Platform (if gateway-related)
No response
Operating System
Ubuntu 24.04.4 LTS (Noble Numbat)
Python Version
Python 3.12.3
Hermes Version
v2026.4.8-461-g4eecaf06
Relevant Logs / Traceback
- - Hermes Agent 版本:最新版本
- 操作系统:Linux
- 模型:Qwen3-235B-A22B (上下文窗口大小:32,768 tokens)
-
Root Cause Analysis (optional)
No response
Proposed Fix (optional)
修改 run_agent.py 文件中的上下文长度检查逻辑,在用户明确设置了 context_length 时跳过最小长度检查:
# Skip minimum context length check if user explicitly set context_length
in config
if _ctx and _ctx < MINIMUM_CONTEXT_LENGTH and self._config_context_length
is None:
raise ValueError(
f"Model {self.model} has a context window of {_ctx:,} tokens, "
f"which is below the minimum {MINIMUM_CONTEXT_LENGTH:,} required "
f"by Hermes Agent. Choose a model with at least "
f"{MINIMUM_CONTEXT_LENGTH // 1000}K context, or set "
f"model.context_length in config.yaml to override."
Are you willing to submit a PR for this?
Bug Description
错误信息提示可以通过设置
model.context_length来覆盖,但实际上这个设置被忽略了。复现步骤
期望行为
当用户在 config.yaml 中明确设置了
model.context_length时,应该尊重用户的设置,绕过最小长度检查。实际行为
即使设置了
model.context_length,仍然会检查这个值是否小于 64,000 tokens,并触发错误。解决方案
修改
run_agent.py文件中的上下文长度检查逻辑,在用户明确设置了context_length时跳过最小长度检查:环境信息
Steps to Reproduce
Model Qwen3-235B-A22B has a context window of 32,768 tokens, which is below the minimum 64,000 required by Hermes Agent. Choose a model with at least 64K context, or set model.context_length in config.yaml to override.
Expected Behavior
当用户在 config.yaml 中明确设置了 model.context_length 时,应该尊重用户的设置,绕过最小长度检查。
Actual Behavior
即使设置了 model.context_length ,仍然会检查这个值是否小于 64,000 tokens,并触发错误
Affected Component
Gateway (Telegram/Discord/Slack/WhatsApp)
Messaging Platform (if gateway-related)
No response
Operating System
Ubuntu 24.04.4 LTS (Noble Numbat)
Python Version
Python 3.12.3
Hermes Version
v2026.4.8-461-g4eecaf06
Relevant Logs / Traceback
Root Cause Analysis (optional)
No response
Proposed Fix (optional)
修改 run_agent.py 文件中的上下文长度检查逻辑,在用户明确设置了 context_length 时跳过最小长度检查:
Are you willing to submit a PR for this?