Issue Station
Sign In
Tools hereToolbox

hermes - ✅(Solved) Fix C:/Program Files/Git/model 选择器中增加思考模式（reasoning）的可视化配置 [1 pull requests, 1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
NousResearch/hermes-agent#25364•Fetched 2026-05-14 03:46:59
View on GitHub
Comments
0
Participants
1
Timeline
6
Reactions
0
Author
ZYC99ZYC99
Participants
ZYC99ZYC99
Timeline (top)
labeled ×4cross-referenced ×2

PR fix notes

PR #25372: feat: model picker favorites/LRU + reasoning sub-picker

Description (problem / solution / changelog)

Summary

Closes #25363, closes #25364

Feature A: Model Favorites & LRU Ordering

  • Added model.recent (LRU, max 10) and model.favorites (pinned) to config
  • Migrated old model: "string" to model: {default: ..., recent: [], favorites: [], reasoning_defaults: {}}
  • Model picker now sorts: favorites (★) first, recents (🕐) second, rest as-is
  • Added /model --favorite <name> and /model --unfavorite <name> commands

Feature B: Reasoning Configuration in Picker

  • Added supports_reasoning(model_id, provider) detecting o1/o3/o4, gpt-5, claude-sonnet-4/opus-4, deepseek-r1, grok-4, gemini-2.5, qwq
  • After selecting a reasoning-capable model, a lightweight sub-picker shows effort levels (none/low/medium/high)
  • Reasoning defaults are persisted per model in model.reasoning_defaults

Files Changed

  • cli.py: +189 lines (picker sorting, LRU, favorites, reasoning stage)
  • hermes_cli/config.py: +23 lines (dict migration, backward compat)
  • hermes_cli/models.py: +53 lines (supports_reasoning function)

Test Plan

  • 101 model-related tests pass (validation, persistence, picker viewport)
  • Manual verification of supports_reasoning() for all known patterns
  • Manual verification of config migration (old string -> new dict, pass-through, empty)

Changed files

  • cli.py (modified, +185/-4)
  • hermes_cli/config.py (modified, +21/-2)
  • hermes_cli/models.py (modified, +53/-0)

Code Example

选择思考模式 (reasoning):
  1. none     — 不思考
  2. low      — 低强度
  3. medium   — 中等（默认）
  4. high     — 高强度

---

model:
  reasoning_defaults:
    "deepseek/deepseek-chat": "medium"
    "anthropic/claude-sonnet-4": "high"
RAW_BUFFERClick to expand / collapse

Feature Description

/model 选择器确认模型后，自动弹出 reasoning 等级选项（如果该模型支持），不必单独执行 /reasoning 命令。

Motivation

目前切换模型后再配置推理等级需要两步操作：/model 选模型 → /reasoning 选等级。对于支持 reasoning 的模型（DeepSeek、Claude thinking、o1/o3 等），这两步应该合并，降低认知负担。

Proposed Solution

在 /model 确认模型后，检测该模型是否支持 reasoning。如果支持，弹出轻量选择：

选择思考模式 (reasoning):
  1. none     — 不思考
  2. low      — 低强度
  3. medium   — 中等（默认）
  4. high     — 高强度

永久记忆方案：config 中支持按模型绑定默认 reasoning 级别：

model:
  reasoning_defaults:
    "deepseek/deepseek-chat": "medium"
    "anthropic/claude-sonnet-4": "high"

Alternatives Considered

  • 保持 /reasoning 独立命令：割裂，需两步
  • 在 /model 列表中嵌套子菜单：可能太过复杂

Scope

Small-Medium — 涉及 /model 命令交互流程 + 模型能力检测（该模型是否 support_reasoning）+ config 持久化。

Additional Context

/reasoning 命令本身工作正常，但缺少可视化引导，新用户可能不知道这个功能存在。

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#ssr#installation#API versioning#request timeout

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING