zai-org / zai-org/feedback

Persistent model picker with per-chat model memory in chat.z.ai

Open
#146 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

priority: P2
Dominant language
No language data
Stars
22
Forks
1
PR merge metrics
No merged PRs in 30d

Description

[Suggestion / Feature] Persistent model picker with per-chat model memory in chat.z.ai

Verified demand: ~478 messages discuss switching models or wishing to stay on a specific model. Users frequently want to switch models mid-session and keep prior context, and to have each chat remember which model it was using.

Proposal:

  1. A persistent model selector that lets users switch models without losing conversation history (extends issue #122 — switch model mid-conversation).
  2. Per-chat model memory: each conversation remembers its model so reopening it resumes on the same one.

Why this matters: During GLM-5.2 saturation (issue #142), users who switch to GLM-5.1 expect that choice to stick — today it can fall back silently. Persistent selection makes model choice explicit and trustworthy.

Regards,
Roman

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Review issue #122 for the existing model-switching context and issue #142 for the saturation fallback behavior. Confirm how chat.z.ai currently selects models and stores conversation state; done means users can switch models without losing history and reopening a chat restores its selected model.

Written by the indexing model from the issue text.

Assessment

Domain
frontend
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.