[octo-patch]

Reason: Add the current MiniMax model as the first selection while preserving legacy and custom model choices.

• Add MiniMax-M3.1-Flash to all global and China OpenAI-compatible and Anthropic-compatible provider settings.
• Configure its context window, multimodal inputs, adaptive thinking, reasoning effort options, and flat token pricing.
• Keep existing legacy and custom entries unchanged and add regression coverage for ordering and metadata.

Validation:

• dotnet test tests/BotSharp.LLM.Tests/BotSharp.LLM.Tests.csproj --configuration Release --filter FullyQualifiedName~MiniMaxProviderTests --nologo (4 passed).
• git diff --check.
• Secret scan passed with the configured credential pattern.

[qodo-code-review]

Code Review by Qodo

Sorry, something went wrong

We weren't able to complete the code review on our side. Please try again manually by commenting /agentic_review on this PR.

[qodo-code-review]

PR Summary by Qodo

Add MiniMax-M3.1-Flash to all MiniMax provider configurations

✨ Enhancement ⚙️ Configuration changes 🧪 Tests 🕐 10-20 Minutes

AI Description

• Add MiniMax-M3.1-Flash as the first choice across four protocol and region configurations.
• Define its multimodal, reasoning, context-window, and flat-pricing metadata while retaining
 existing models.
• Extend configuration tests to verify model ordering and Flash metadata.

Diagram

graph TD
  C["WebStarter config"] --> S["LLM settings"] --> OG["OpenAI global"]
  S --> OC["OpenAI China"]
  S --> AG["Anthropic global"]
  S --> AC["Anthropic China"]

Loading

High-Level Assessment

Adding entries to the existing provider lists fits the configuration-driven model selection design. Sharing the repeated metadata through a new configuration mechanism would add complexity for this single model.

Files changed (2) +157 / -1

Tests (1) +13 / -1

MiniMaxProviderTests.csVerify Flash ordering and configuration metadata +13/-1

Verify Flash ordering and configuration metadata

• Extends the WebStarter configuration test to assert Flash is first for all four aliases and checks its context, modalities, reasoning defaults and options, and flat pricing. Existing M3 and M2.7 assertions remain in place.

tests/BotSharp.LLM.Tests/MiniMaxProviderTests.cs

Other (1) +144 / -0

appsettings.jsonConfigure Flash for four MiniMax provider aliases +144/-0

Configure Flash for four MiniMax provider aliases

• Prepends MiniMax-M3.1-Flash to the global and China OpenAI-compatible and Anthropic-compatible model lists. Defines a one-million-token context window, text/image/video inputs, adaptive thinking, effort options, and flat token costs without changing the existing model entries.

src/WebStarter/appsettings.json