Skip to content

feat: add MiniMax as LLM provider with model registry and cost tracking#5642

Open
octo-patch wants to merge 1 commit intoHelicone:mainfrom
octo-patch:feature/add-minimax-provider
Open

feat: add MiniMax as LLM provider with model registry and cost tracking#5642
octo-patch wants to merge 1 commit intoHelicone:mainfrom
octo-patch:feature/add-minimax-provider

Conversation

@octo-patch
Copy link
Copy Markdown

Summary

  • Add MiniMax as a first-class LLM provider in Helicone's model registry
  • Register 4 models: MiniMax-M2.7, M2.7-highspeed, M2.5, M2.5-highspeed with accurate pricing
  • Integrate with gateway routing (URL pattern matching, provider name mapping, usage processing)
  • Full test coverage: 24 unit tests + 8 integration tests (all passing)

Changes

Provider (packages/cost/models/providers/)

  • minimax.tsMiniMaxProvider extending BaseProvider, OpenAI-compatible (api.minimax.io/v1/chat/completions)
  • index.ts — Register singleton, add to ResponsesAPIEnabledProviders
  • priorities.ts — Priority 4 (standard third-party)

Model Registry (packages/cost/models/authors/minimax/)

  • minimax-m2.7/ — M2.7 ($0.20/$1.10 per 1M, 1M ctx) + M2.7-highspeed ($0.10/$0.55 per 1M)
  • minimax-m2.5/ — M2.5 ($0.20/$1.10 per 1M, 204K ctx) + M2.5-highspeed ($0.10/$0.55 per 1M)
  • index.ts / metadata.ts — Aggregation and author metadata

Integration

  • registry.ts / registry-types.ts — Add MiniMax models and endpoints to combined registry
  • provider-helpers.tsheliconeProviderToModelProviderName("MINIMAX") + dbProviderToProvider
  • mappings.ts — URL pattern api.minimax.ioMINIMAX provider name
  • types.ts — Add "minimax" to AUTHORS, "MiniMax" to Tokenizer
  • getUsageProcessor.ts — Route to OpenAIUsageProcessor (OpenAI-compatible format)

Tests (packages/__tests__/cost/providers/)

  • minimax.test.ts — 24 unit tests (provider class, registry, name mapping, endpoints, pricing, parameters)
  • minimax-integration.test.ts — 8 integration tests (model parsing, URL building, usage processing, PTB endpoints, full request flow)

Test plan

  • All 222 existing + new tests pass (npx jest __tests__/cost/ --no-coverage)
  • Registry snapshots updated
  • No breaking changes to existing providers

Add MiniMax as a first-class LLM provider in Helicone's model registry,
enabling cost tracking, usage processing, and gateway routing for
MiniMax-M2.7, M2.7-highspeed, M2.5, and M2.5-highspeed models.
Copy link
Copy Markdown
Contributor

@greptile-apps greptile-apps bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.

@vercel
Copy link
Copy Markdown

vercel bot commented Mar 21, 2026

@octo-patch is attempting to deploy a commit to the Helicone Team on Vercel.

A member of the Team first needs to authorize it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant