[ENHANCEMENT] Add MiniMax M2.7 to Minimax Provider

[DarkEden-coding]

Problem (one or two sentences)

The new minimax m2.7 model is not in the current minimax provider.

The model has 204,800 token context with:
$0.30 | $1.20 | $0.06

base pricing (input, output, cache read)

Context (who is affected and when)

Everyone who uses the minimax provider

Desired behavior (conceptual, not technical)

Have the m2.7 model available

Constraints / preferences (optional)

No response

Request checklist

• I've searched existing Issues and Discussions for duplicates
• This describes a specific problem with clear context and impact

Roo Code Task Links (optional)

No response

Acceptance criteria (optional)

No response

Proposed approach (optional)

No response

Trade-offs / risks (optional)

No response

[roomote-v0]

I've analyzed this issue and here's my proposed implementation plan:

The MiniMax M2.7 model needs to be added to the existing MiniMax provider. Based on the current model definitions in packages/types/src/providers/minimax.ts, I'll add a new "MiniMax-M2.7" entry with the following configuration:

1. Add model entry to packages/types/src/providers/minimax.ts with:

   • contextWindow: 204_800 (as specified in the issue)
   • maxTokens: 16_384 (consistent with other MiniMax models)
   • inputPrice: 0.3, outputPrice: 1.2, cacheReadsPrice: 0.06 (as specified: $0.30 / $1.20 / $0.06)
   • cacheWritesPrice: 0.375 (consistent with other MiniMax models -- will confirm if different)
   • supportsPromptCache: true, supportsImages: false, preserveReasoning: true
   • includedTools: ["search_and_replace"], excludedTools: ["apply_diff"] (consistent with existing models)

2. Update the default model to "MiniMax-M2.7" since it's the newest model (or keep M2.5 as default -- open to guidance here).

3. Add test coverage in src/api/providers/__tests__/minimax.spec.ts for the new model entry.

One question: should the cache writes price remain at $0.375 (matching other MiniMax models), or is there a different price for M2.7? Also, should M2.7 become the new default model for the MiniMax provider?

Please mention @roomote in a comment saying that you approve this plan, and I'll proceed with the implementation.

[DarkEden-coding]

plan approved with following info:

Model	Input	Output	Prompt caching Read	Prompt caching Write
MiniMax-M2.7	$0.3 / M tokens	$1.2 / M tokens	$0.06 / M tokens	$0.375 / M tokens

keep default to 2.5 as 2.7 is not tested

@roomote

[roomote-v0]

I've seen your comment and the approved plan. I'll proceed with adding MiniMax-M2.7 to the MiniMax provider with the specified pricing, keeping M2.5 as default. Will open a PR shortly.

[roomote-v0]

Opened PR #11954 with the MiniMax-M2.7 model addition. Default remains M2.5 as requested. All tests pass. Feedback welcome! #11954

[alanchenchen]

What about MiniMax-M2.7-highspeed？I wanna add custom model, but how can do that @roomote

[DarkEden-coding]

@roomote
It seems that custom model selection is not currently working, it just defaults to minimax2.5, please investigate if this is an actual issue

[DarkEden-coding]

@alanchenchen seems that many providers have bugs with custom model selection at the moment #11965 #11964

[alanchenchen]

@DarkEden-coding yeah, i have switch to use OpenAI Compatible provider, custom model works. Maybe you could do that.😎