[FEATURE] Per-message model selection in Claude Code CLI

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

I often work with Claude Code using Opus 4.8 as my default model for complex tasks. However, for small operational requests (e.g., "describe this repository briefly", code formatting, simple refactoring), I would like to use a cheaper model like Haiku without switching the entire session.

Currently, I have to either:

• Manually switch models with /model (affects all subsequent messages)
• Use separate terminal sessions (cumbersome)

This prevents tactical cost optimization for mixed workflows.

Proposed Solution

Add syntax to specify a model for a single message without affecting the session default:

describe this repository briefly @haiku

or

/describe this repository briefly --model haiku

This would allow:

• Using Opus 4.8 for complex reasoning tasks
• Using Haiku for simple operational tasks
• No need to manage multiple sessions
• Better cost optimization (60-80% savings on per-message basis)

The main conversation context and history should remain unchanged, and the default model should persist for subsequent messages.

Alternative Solutions

_No response_

Priority

Medium - Would be very helpful

Feature Category

CLI commands and flags

Use Case Example

_No response_

Additional Context

_No response_