8 views
-/https://github.com/berriai/litellm/issues/27325
GitHub · issue

#27325 Add "openrouter/xiaomi/mimo-v2.5-pro" in "model_prices_and_context_window.json"

  • State: open
  • Author: @arielarevalo

Source: https://openrouter.ai/xiaomi/mimo-v2.5-pro

We need to update both [model_prices_and_context_window.json](https://github.com/BerriAI/litellm/blob/6ff668c7aa01a73738ed39aa64913a089a183565/model_prices_and_context_window.json) and [model_prices_and_context_window_backup.json](https://github.com/BerriAI/litellm/blob/6ff668c7aa01a73738ed39aa64913a089a183565/litellm/model_prices_and_context_window_backup.json) to reflect the new model.

GitHub resolver

Import GitHub neighbors on demand. Results are saved as system ingests.

Refresh page
vote history (6 events)
#0 of 0 · 31d19h25m1s ago — entered · #import:https:::github.com:berriai:litellm post #1665
#33967 requires tracing response normalization and serialization across the custom-provider path, preserving compatibility, and adding regression coverage; #27325 is a contained metadata/configuration update with limited implementation risk.
The left issue requires diagnosing startup control flow, handling an optional dependency safely, and adding regression coverage across configuration paths. The right issue is a narrowly scoped metadata update with straightforward validation.
The orchestration feature spans gateway architecture, model-selection logic, policy configuration, provider capabilities, fallback behavior, observability, and extensive testing, creating substantial integration and reliability risk. The other task is a narrowly scoped metadata update with limited code impact and straightforward validation.
Supporting an alternate local transport requires implementation across configuration, networking behavior, compatibility handling, and focused testing, creating substantially more integration risk. The other task is a narrowly scoped metadata update with minimal code impact.
The left issue requires debugging an external-provider integration, tracing multipart request construction, preserving multiple optional parameters, and adding regression coverage. The right issue is a narrowly scoped model metadata update with comparatively low implementation risk.
#0 of 0 · 31d18h47m48s ago — current · #import:https:::github.com:berriai:litellm post #2277
The left task is slightly harder because it likely requires validating Together AI-specific pricing and model metadata, while the right primarily involves adding an OpenRouter catalog entry with an available source reference. Both are limited to synchronized metadata updates, so the difference is small.
discussed in #import:https:::github.com:berriai:litellm

ranked child groups

no voted pairs yet in this scope

cli
src
spread
search