Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
app: source featured models from experimental recommendations endpoint (#15909)
938ca6e Bruce MacDonald committed 6 months ago
fix: resolve OpenClaw gateway launch timeout on Windows by enforcing IPv4 loopback (#15726)
8f39fff Pratham Agarwal committed 6 months ago
metal: harden for ggml initialization failures (#15755)
4fe5609 Daniel Hiltgen committed 6 months ago
app: remove ollama update url env var used for testing (#15905)
917324b Bruce MacDonald committed 6 months ago
launch: use vram bytes for model recommendations (#15885)
b6447ca Parth Sareen committed 6 months ago
server/launch: add model recommendations cache endpoint (#15868)
321cc8a Parth Sareen committed 6 months ago
mlxrunner: decouple models from attention cache storage layout
2bbe240 Jesse Gross committed 7 months ago
app/server: fix desktop app startup killing active `ollama launch` sessions (#15657)
3cab8a7 Eva H committed 6 months ago
mlx: Support NVIDIA TensorRT Model Optimizer import (#15566)
03aee88 Daniel Hiltgen committed 6 months ago
tokenizer: fix multi-regex BPE offset handling (#15844)
ec9b4e9 Daniel Hiltgen committed 6 months ago
mlxrunner: track sampler history in a fixed-size ring buffer
30f86cb Jesse Gross committed 7 months ago
openai: map responses reasoning effort to think (#15789)
ea01af6 Parth Sareen committed 7 months ago
launch: resave managed integration config when live config drifts (#15776)
b4442c6 Eva H committed 7 months ago
launch: keep launch recommended models in a fixed canonical order (#15750)
85ff8e4 Eva H committed 7 months ago