Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
app/ui: handle unspecified bind addresses and wait for server in ollama proxy (#13159)
8dbc9e7 Eva H committed 11 months ago
model: default gemma 3 rope scale to 1.0, apply corrections based on layer counts (#13453)
4ff8a69 Jeffrey Morgan committed 11 months ago
model: fix global layer rope scale values for gemma 3 (#13452)
1b308e1 Jeffrey Morgan committed 11 months ago
docs: add docs for v1/responses and rework openai compat section (#13416)
9f78228 Devon Rifkin committed 11 months ago
openai: add tool call appending to previous assistant message (#13434)
9b2035d Parth Sareen committed 11 months ago
model: fix rotary embeddings for ministral 3 (#13432)
2dfb744 Jeffrey Morgan committed 11 months ago
model: conversion and hyperparameter fixes for ministral and devstral (#13424)
a838421 Jeffrey Morgan committed 11 months ago
cmd/bench: fix options table in cmd/bench/README.md (#13216)
56b8fb0 Julia Scheaffer committed 11 months ago
feat: llama.cpp bump (17f7f4) for SSM performance improvements (#13408)
b956930 Gabe Goodhart committed 11 months ago
app/ui: use requestAnimationFrame to prevent bottom line cutoff in streaming thinking display (#13137)
c34fc64 Eva H committed 11 months ago
app/ui: refactor to use Ollama endpoints for user auth and health checks (#13081)
7cf6f18 Eva H committed 11 months ago
app/ui: fix model capabilities not updating after download completion (#13179)
bbbb6b2 Eva H committed 11 months ago