| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
Sorry, something went wrong.
- Auto-detect MiniMax when MINIMAX_API_KEY is set (no LLM_PROVIDER needed) - Built-in defaults: model (MiniMax-M2.5), base URL, temperature range - Temperature clamping to MiniMax's accepted (0.01, 1.0] range - Custom model/base URL overrides via MINIMAX_MODEL / MINIMAX_BASE_URL - 15 unit tests + 3 live integration tests - Updated .env.sample and README with MiniMax configuration
- Set MiniMax-M2.7 as default model (was M2.5) - Update highspeed model reference to MiniMax-M2.7-highspeed - Keep all previous models as available alternatives via env var - Update related unit tests and documentation
- Upgrade default model from MiniMax-M2.7 to MiniMax-M3 - Keep MiniMax-M2.7-highspeed as alternative - Update related tests, README, and .env.sample - All 15 unit tests pass
| Back | FazBrowse Home | New Git URL |
Summary
Upgrade the MiniMax provider's default model from M2.7 to M3, while keeping M2.7 and M2.7-highspeed available as alternatives.
Changes
Why
MiniMax M3 is the latest generation model with a 512K context window, up to 128K output, and image input support. Bumping the default gives users a more capable out-of-the-box experience.
MiniMax-M2.7 and MiniMax-M2.7-highspeed are still selectable via the MINIMAX_MODEL env var for users who prefer them.
Testing