Findings [1] 2026-09-27 v1.103.0 fix(proxy): unregister logging callbacks removed from the stored conf… [2] 2026-09-27 1.64.1.dev20260926 Streamlit nightly 1.64.1.dev20260926
Where Kimbodo Comes In Kimbodo builds and operates this in production for businesses — see our AI Application Development practice, or Estimate My AI Application.
Sources
[1] v1.103.0 [2] 1.64.1.dev20260926
Findings [1] 2026-09-26 1.64.1.dev20260925 Streamlit nightly 1.64.1.dev20260925
Where Kimbodo Comes In Kimbodo builds and operates this in production for businesses — see our AI Application Development practice, or Estimate My AI Application.
Sources
[1] 1.64.1.dev20260925
Findings [1] 2026-09-25 v0.40.0 What's Changed Models run on MLX on Apple Silicon by default In this release, on Apple Silicon devices, model architectures supported by the MLX runtime will automatically run on MLX. ollama pull qwen3.8 ollama run qwen3.8 During the pre-release we will be testing and enabling additional models. Full Changelog: v0.34.4...v0.40.0-rc0…
Findings [1] 2026-09-24 langchain-core==1.6.5 Changes since langchain-core==1.6.4 release(core): 1.6.5 (#40816) fix(core): abbreviate long tool IDs in XML buffer strings (#40792) [2] 2026-09-24 langchain-openai==1.6.6 Changes since langchain-openai==1.6.5 release(openai): 1.6.6 (#40800) fix(openai): raise on error events in stream path (#40791) [3] 2026-09-24 v0.34.4 What's Changed Structured outputs on thinking models now apply…
Findings [1] 2026-09-24 v0.34.4 What's Changed server: fix intermittent "model not found" errors. by @rick-github in #18438 server: apply structured outputs in a single pass on thinking models by @jessegross in #18479 app: avoid System Events for ChatGPT/Codex detection by @hoyyeva in #18601 llama.cpp: version update by @dhiltgen in #18577 MLX: version bump by…
Findings [1] 2026-09-23 v0.34.4-rc0: mlx: speed up Qwen 3.8 prompt processing (#18550) mlx: speed up Qwen 3.8 prompt processing Use MLX's gated-delta kernel for long scans and fold dense MLP global scales into SwiGLU. address comments [2] 2026-09-22 langchain-openai==1.6.4 Changes since langchain-openai==1.6.3 release(openai): 1.6.4 (#40775) chore(model-profiles): refresh openai model profile data (#40774)…
Findings [1] 2026-09-21 langchain-openai==1.6.3 Changes since langchain-openai==1.6.2 release(openai): 1.6.3 (#40719) fix(openai): expose inferred Responses API routing at initialization (#40715) chore(deps): bump anyio from 4.11.0 to 4.14.2 in /libs/partners/openai (#40629) fix(openai): support GPT-6 request constraints (#40443) [2] 2026-09-21 v4.7.0a2 4.7.0a2 (Full Changelog) Security fixes GHSA-3325-v43h-43rv - moderate GHSA-6966-vjj6-99xv - high GHSA-jwrc-gm9j-263p - moderate…
What Happened
Three relevant release updates surfaced across AI/ML open-source components this cycle:
langchain-typesafe published initial iterations (0.0.1a2 → 0.0.1a3). Highlights include a new TypeSafeClassifier, invocation-scoped classifier questions, experimental middleware (AutoModeMiddleware, ModelRouterMiddleware), and a metadata trace bugfix [1].
LiteLLM published v1.103.0-rc.1 with signed Docker images (cosign), many new provider integrations and…
What Happened
Two relevant releases surfaced that teams maintaining AI/ML apps should act on:
Ollama v0.34.3: API change — GET /api/show now includes a model's "thinking" control options and default (example payload: {"thinking": {"values": ["low","high","max"], "default":"max"}}). New support enables Nemotron H vision models to run on Apple Silicon via MLX. macOS app behavior…
What Happened
Ollama v0.34.3 — API change: GET /api/show now advertises per-model "thinking" controls and defaults; Apple Silicon (MLX) support for Nemotron H vision models; macOS app no longer reopens closed windows on activation [1].
LangChain 1.4.2 — Patch bump with an important fix to preserve model-generated tool calls in human-in-the-loop…
What Happened
A set of targeted releases and prereleases across model-serving and ML tooling introduced new features and one high-impact memory fix, plus several experimental library releases that require cautious adoption:
ollama v0.34.2: added a first-run setup flow (CLI options to sign in or continue locally) with desktop-app sync on macOS/Windows, an ollama://apps…
What Happened
LiteLLM (v1.102.0-rc.2 → v1.103.0-dev.1)
All official LiteLLM Docker images are now signed with cosign; the project publishes a committed public key you can pin for verification. Example pinned verification is provided in the release notes [2][5].
v1.102.0-rc.2 backports request-param leak fixes (security/privilege leak patches) into the RC branch (PRs…