Model Drop — Issue #001

New AI models & tool drops, every week

Week of: 2026-10-15 (America/Chicago pipeline run)

Sources checked: 54/55

Items after QA filter: 78 (of 284 raw new)

Draft generated from live fetches. Every bullet links to the official source. No fabricated releases.

Highlights

• Anthropic Python SDK — v1.15.0 (2026-10-15)
• Ollama — v0.40.5 (2026-10-15)
• Vercel AI SDK — @ai-sdk/[email protected] (2026-10-15)
• Vercel AI SDK — @ai-sdk/[email protected] (2026-10-15)
• Vercel AI SDK — @ai-sdk/[email protected] (2026-10-15)
• llama.cpp — v0.7.0 (2026-10-15)
• ComfyUI — v0.41.1 (2026-10-15)
• llama.cpp — b11669 (2026-10-15)

By category

AI coding tools

• Claude Code: Claude Code v2.1.300 — 2026-10-14
• Claude Code 2.1.300 brings smarter prompt handling with virtual text that won't interfere with your submissions, plus new ways to customize agent views and control advisor tools directly. You can now rewrite workflow agent prompts on the fly and track detailed cache performanc...
• Model or capability update — may change quality, latency, or cost.
• Gemini CLI: Gemini CLI v0.64.0 — 2026-10-13
• Gemini CLI v0.64.0 brings smoother performance with fixes that prevent CPU hangs and improve how your settings are saved and recovered from crashes. We've also made file operations more reliable and fixed several stability issues to give you a better experience overall. Check ...
• Model or capability update — may change quality, latency, or cost.
• Cursor: Oct 13, 2026 — 2026-10-13
• Oct 13, 2026 · Changelog
• Affects day-to-day AI coding workflows.
• OpenAI Codex CLI: OpenAI Codex CLI rust-v0.162.1 — 2026-10-09
• OpenAI Codex CLI rust-v0.162.1 fixes a critical crash when handling multi-line questions and resolves startup issues caused by feature setting mismatches. These updates make the tool more reliable for everyday coding tasks. Check the full changelog to see what else improved un...
• Affects day-to-day AI coding workflows.

APIs

• Gemini API: overview page — 2026-10-13
• API/SDK change — skim if you integrate this provider.
• Gemini API: Gemini models — 2026-10-13
• Model or capability update — may change quality, latency, or cost.
• Gemini API: the API versions topic — 2026-10-13
• API/SDK change — skim if you integrate this provider.

Agents

• CrewAI: 1.15.28 — 2026-10-14
• What's Changed Features Add crewai eval command to report where a stopped model broke. Use LiteLLM catalog for context windows. Bug Fixes Fix flow to learn from resumed human feedback and preserve reviewed output types. Warn when non-empty apps resolve to zero platform tools. ...
• Product update — skim if you use this tool.
• CrewAI: 1.15.27 — 2026-10-09
• What's Changed Features Add XPU to OpenCLIP provider device options Add deepinfra as an OpenAI-compatible provider Record why an evaluation stopped before its verdict Record what the run app's Deploy met before an attempt Implement per-run cost and time tracking in crewai eval...
• Product update — skim if you use this tool.
• CrewAI: 1.15.26 — 2026-10-08
• What's Changed Bug Fixes Fix ownership retention of rwlock when acquisition is interrupted. Resolve issue with documentation site source URLs not being preserved. Fix task output display in log-tasks-outputs to show actual output. Documentation Update snapshot and changelog fo...
• Product update — skim if you use this tool.

Audio

• ElevenLabs: Introducing Instant Soundtracks — 2026-10-15
• Introducing Instant Soundtracks, convert video to music and sound effects automatically.
• Product update — skim if you use this tool.
• ElevenLabs: Audio deepfake detection: How it works & how to verify ElevenLabs audio — 2026-10-14
• Learn how audio deepfake detection works, how watermarking differs from statistical classifiers, and how to check whether audio came from ElevenLabs.
• Product update — skim if you use this tool.
• ElevenLabs: How we solve 22,000 support tickets every month with ElevenAgents — 2026-10-14
• We utilize ElevenAgents procedures, simulation tests, tools, and more to enable the agent to look-up account information, take actions, and escalate
• Product update — skim if you use this tool.

Automation

• n8n: [email protected] — 2026-10-15
• 2.43.6 (2026-10-15) Bug Fixes core: Create an XML parser for each webhook request body ( #41113 ) ( ea3ee53 )
• Product update — skim if you use this tool.
• n8n: stable — 2026-10-15
• 2.43.6 (2026-10-15) Bug Fixes core: Create an XML parser for each webhook request body ( #41113 ) ( ea3ee53 )
• Product update — skim if you use this tool.
• n8n: [email protected] — 2026-10-15
• 2.44.2 (2026-10-15) Bug Fixes core: Create an XML parser for each webhook request body ( #41112 ) ( c52f181 )
• Product update — skim if you use this tool.

Data & browsing

• Browserbase: Parallel is now available in Browserbase Search — 2026-10-14
• We’re excited to welcome a new leading search provider to our Browserbase Search API.
• Product update — skim if you use this tool.

Frameworks

• LlamaIndex: ExtractBench Pro: The Long Tail of Document Extraction — 2026-10-15
• ExtractBench Pro: The Long Tail of Document Extraction
• Product update — skim if you use this tool.
• DSPy: 3.4.1 — 2026-10-12
• DSPy 3.4.1 DSPy 3.4.1 makes GEPA's structured InstructionProposer the default instruction proposer and lets dspy.RLM build DSPy sub-agents from inside its REPL. It also adds Python 3.15 support, fixes a 3.4.0 import crash, and includes streaming, adapter, and interpreter fixes...
• Product update — skim if you use this tool.

Frontier labs & models

• Google DeepMind: Accelerating the search for antimicrobial peptides with Google Antigravity — 2026-10-15
• Accelerating the search for novel antimicrobial peptides using Google Antigravity for Science.
• Product update — skim if you use this tool.
• OpenAI: Teaching students to think and learn through questions — 2026-10-15
• Celebrate Teachers’ Day with AI for Educators in Brazilian Portuguese and see how Study Mode can help teachers support active, step-by-step learning.
• Product update — skim if you use this tool.
• Anthropic: Oct 15, 2026 Announcements An update on independent evaluators — 2026-10-15
• Product update — skim if you use this tool.
• Mistral AI: Research Introducing Timestral State-of-the-art probabilistic forecasts with just 209M parameters, outperforming larger models on CPU. October 15, 2026 By Mistral — 2026-10-15
• Research
• Model or capability update — may change quality, latency, or cost.
• OpenAI: Amp puts AI agents to work beyond code with OpenAI models — 2026-10-14
• See how Amp pushes the frontier of what agents can do with GPT-6 Astra, Ultrafast, GPT-Live-1, and Sign in with ChatGPT.
• Model or capability update — may change quality, latency, or cost.
• OpenAI: SentinelOne cuts malware analysis from weeks to hours — 2026-10-14
• With OpenAI Daybreak, SentinelOne cuts weeks of malware analysis to hours, creating capacity to investigate more threats.
• Product update — skim if you use this tool.
• Google DeepMind: Google DeepMind and Singapore Government collaborate to build fine-tuned AI model for scam detection — 2026-10-14
• Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.
• Model or capability update — may change quality, latency, or cost.
• Anthropic: Oct 14, 2026 Policy Preparing for the 2026 US midterms — 2026-10-14
• Product update — skim if you use this tool.
• Anthropic: Oct 14, 2026 Announcements Anthropic opens Singapore office and signs MOU with the Government of Singapore — 2026-10-14
• Product update — skim if you use this tool.
• Google DeepMind: Developing Gemini 4 Argon responsibly and safely — 2026-10-13
• Model or capability update — may change quality, latency, or cost.
• Google AI: Gemini 4 Argon: Built to solve your hardest problems — 2026-10-13
• TBD
• Model or capability update — may change quality, latency, or cost.
• xAI: Oct 14, 2026 Introducing Grok Voice 3.0 The world’s most intelligent voice model. Read More — 2026-10-12
• Oct 14, 2026
• Model or capability update — may change quality, latency, or cost.
• xAI: Product · Oct 12, 2026 Introducing The X Subscription — 2026-10-12
• Product update — skim if you use this tool.
• xAI: Product · Sep 28, 2026 Team Bots: shared AI teammates that learn as they work — 2026-10-12
• Product update — skim if you use this tool.

Image / video / media

• ComfyUI: v0.41.1 — 2026-10-15
• ComfyUI v0.41.1
• Product update — skim if you use this tool.
• ComfyUI: v0.41.0 — 2026-10-15
• ComfyUI v0.41.0
• Product update — skim if you use this tool.
• ComfyUI: v0.40.0 — 2026-10-13
• What's Changed [Server.py] Update CSP to allow blob: and data: by @S0yKaf in #15565 Add comfy_attention and AttentionTensorContainer support to rest of models by @comfyanonymous in #16806 Implement the asset export API locally (CORE-454) by @jtydhr88 in #16578 Report 3D models...
• Product update — skim if you use this tool.

Inference & infra

• llama.cpp: v0.7.0 — 2026-10-15
• Overview llama.cpp v0.7.0 adds the LiliCorr speculative decoding technique and MTP for GLM5-Next, a new GPU cache for MoE experts kept in host memory ( --moe-cache-mib , up to ~2.2x decode speedup when experts are offloaded), and a large batch of new models: K2 Horizon (dense ...
• Product update — skim if you use this tool.
• llama.cpp: b11669 — 2026-10-15
• llama.cpp : bump version to 0.7.0 ( #30603 ) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/56338993 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework Lin...
• Product update — skim if you use this tool.
• llama.cpp: b11662 — 2026-10-15
• sync : ggml Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/56327800 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework Linux: Ubuntu x64 (CPU) Ubuntu arm64...
• Product update — skim if you use this tool.
• LiteLLM: v1.106.0-rc.4 — 2026-10-15
• Verify Docker Image Signature All LiteLLM Docker images are signed with cosign . Every release is signed with the same key introduced in commit 0112e53 . Verify using the pinned commit hash (recommended): A commit hash is cryptographically immutable, so this is the strongest w...
• Product update — skim if you use this tool.
• vLLM: v0.32.0rc2 — 2026-10-15
• [CI] Pin DeepSeek-V4.1-Flash-NVFP4 eval to its FP8-Engram revision (#…
• Product update — skim if you use this tool.
• LiteLLM: v1.107.0-dev.2 — 2026-10-15
• Verify Docker Image Signature All LiteLLM Docker images are signed with cosign . Every release is signed with the same key introduced in commit 0112e53 . Verify using the pinned commit hash (recommended): A commit hash is cryptographically immutable, so this is the strongest w...
• Product update — skim if you use this tool.
• vLLM: proto-v0.6.0 — 2026-10-14
• vllm-proto 0.6.0
• Product update — skim if you use this tool.
• [vLLM: v0.32.0rc1: [Dependency] Update Humming and remove NVFP4 scale workaround (#61556)](https://github.com/vllm-project/vllm/releases/tag/v0.32.0rc1) — 2026-10-14
• Signed-off-by: jinzhen.ljz [email protected] Co-authored-by: Codex [email protected] Co-authored-by: Misha Goin [email protected] Co-authored-by: Kevin H. Luu [email protected]
• Possible breaking change — read the notes before upgrading.
• LiteLLM: v1.106.0-rc.3 — 2026-10-14
• Verify Docker Image Signature All LiteLLM Docker images are signed with cosign . Every release is signed with the same key introduced in commit 0112e53 . Verify using the pinned commit hash (recommended): A commit hash is cryptographically immutable, so this is the strongest w...
• Product update — skim if you use this tool.

Local / open models

• Ollama: v0.40.5 — 2026-10-15
• NVIDIA Nemotron 3.5 Super NVIDIA Nemotron 3.5 Super is a vision-language model for coding, reasoning, tool use, and document understanding. It pairs a 120B-parameter hybrid Mamba-2/Transformer mixture-of-experts backbone (12B active per token) with a C-RADIO vision encoder, an...
• Product update — skim if you use this tool.
• Ollama: v0.40.5-rc2 — 2026-10-15
• llama.cpp: version bump b11669 ( #19032 )
• Product update — skim if you use this tool.
• Ollama: v0.40.5-rc1: New model... (#19030) — 2026-10-15
• ci: shorten Windows Vulkan build path Keep the nested shader-generator compiler probes below Windows path limits in checkouts with longer repository names. mlx: fix Nemotron bugs and support global scales Accept published Nemotron layer names while retaining legacy aliases, an...
• Model or capability update — may change quality, latency, or cost.
• Open WebUI: v0.12.2 — 2026-10-14
• Added 📑 MinerU 4.x servers. The MinerU document engine now works with self-hosted MinerU 4.x servers, where every document failed with a 404 because 4.x removed the old parsing interface; 3.x servers work as before, 4.x servers take a tier (flash, basic, standard or advanced) ...
• Product update — skim if you use this tool.
• Open WebUI: v0.12.1 — 2026-10-14
• Added 🧱 External apps in channels. Administrators can now turn on Apps in Admin Settings > General, or with "ENABLE_APPS", and connect external apps in the new Apps tab of Admin Settings by their HTTPS address, reviewing the permissions each app asks for, such as reading its m...
• Product update — skim if you use this tool.
• LM Studio: Bionic 1.1.8 — 2026-10-12
• Product update — skim if you use this tool.
• Open WebUI: v0.12.0 — 2026-10-10
• Added ☎️ Realtime voice calls. Administrators can now switch Call mode in the Voice calls section of Audio settings from Standard to Realtime, so Voice mode talks through an OpenAI Realtime voice model (gpt-realtime-2.1-mini with the Marin voice by default) that handles small ...
• Product update — skim if you use this tool.

Notebooks

• Deepnote: October 9, 2026 — 2026-10-09
• Product update — skim if you use this tool.

Platforms

• Hugging Face: Apertus 1.5 Release Notes — 2026-10-15
• Product update — skim if you use this tool.
• Hugging Face: Welcome Muse Spark Open: open-weight frontier multimodal mixture-of-experts model from Meta — 2026-10-15
• Model or capability update — may change quality, latency, or cost.
• Hugging Face: Welcome FLUX.3 Image Klein in Diffusers — 2026-10-15
• Product update — skim if you use this tool.
• Groq: Partnership Meta's Muse Spark open-weight release Is Now in Limited Preview on GroqCloud October 15, 2026 — 2026-10-15
• Product update — skim if you use this tool.
• AnythingLLM: AnythingLLM v1.18.0 — 2026-10-13
• What's Changed fix: point k8s probes at the server's /api/ping by @u00dxk2 in #6592 fix: don't crash when removing a parsed file while an image is attached by @marmar9615-cloud in #6585 fix: open the clicked image when attachments share a file name by @marmar9615-cloud in #658...
• Product update — skim if you use this tool.

Product analytics

• PostHog: October 15, 2026 — 2026-10-15
• October 15, 2026 · Conversations Team · MCP
• Product update — skim if you use this tool.
• PostHog: October 15, 2026 — 2026-10-15
• October 15, 2026 · Self-Driving Team · Self-driving
• Product update — skim if you use this tool.
• PostHog: October 15, 2026 — 2026-10-15
• October 15, 2026 · Growth Team · Uncategorized
• Product update — skim if you use this tool.

SDKs

• Anthropic Python SDK: v1.15.0 — 2026-10-15
• Features api: add between_tools to thinking types in model capabilities api: add the speed parameter and usage.speed to Messages api: return a Console organization's usage-tier spend cap as a read-only spend limit Chores ci: speed up breaking change detection docs: describe he...
• API/SDK change — skim if you integrate this provider.
• Vercel AI SDK: @ai-sdk/[email protected] — 2026-10-15
• Patch Changes Updated dependencies [ 739e9f6 ] @ai-sdk/[email protected] @ai-sdk/[email protected] @ai-sdk/[email protected]
• API/SDK change — skim if you integrate this provider.
• Vercel AI SDK: @ai-sdk/[email protected] — 2026-10-15
• Patch Changes Updated dependencies [ 739e9f6 ] @ai-sdk/[email protected] @ai-sdk/[email protected] @ai-sdk/[email protected]
• API/SDK change — skim if you integrate this provider.
• Vercel AI SDK: @ai-sdk/[email protected] — 2026-10-15
• Patch Changes Updated dependencies [ 130e871 ] Updated dependencies [ d6ed4b9 ] @ai-sdk/[email protected] @ai-sdk/[email protected] @ai-sdk/[email protected]
• API/SDK change — skim if you integrate this provider.
• Anthropic Python SDK: v1.14.0 — 2026-10-15
• Features api: add execution identity to Managed Agents agents api: add organization type not supported error to spend limit lists api: add telemetry destination test endpoint api: add telemetry destinations to the Admin API client: add native support for OpenTelemetry tracing ...
• API/SDK change — skim if you integrate this provider.
• OpenAI Python SDK: v3.31.0 — 2026-10-14
• 3.31.0 (2026-10-14) Features add browser and computer toolsets (1/4) ( #4073 ) ( 47b9e4e )
• API/SDK change — skim if you integrate this provider.
• OpenAI Python SDK: v3.30.0 — 2026-10-14
• 3.30.0 (2026-10-14) Features api: add beta session spending limits ( #4058 ) ( b570279 ) api: add OCI external-storage provider support ( #4060 ) ( 9eb83fe ) api: add organization Decisions usage reporting ( #4062 ) ( 5e7a23e ) api: expose image tool error and moderation detai...
• API/SDK change — skim if you integrate this provider.
• OpenAI Python SDK: v3.29.0 — 2026-10-11
• 3.29.0 (2026-10-11) Features api: add billing not active session error code ( #4052 ) ( c88b320 ) api: add shida to named Live voices ( #4053 ) ( 7352cf5 ) api: add vault updates and metadata filters ( #4047 ) ( e72f493 ) api: align beta pagination limits and non-null paramete...
• API/SDK change — skim if you integrate this provider.
• Anthropic Python SDK: v1.13.0 — 2026-10-09
• Features api: add types for the Chat and Cowork unified analytics metrics api: add workflows, multiagent configuration and thread status filtering to Managed Agents Chores internal: name the default API version header value
• API/SDK change — skim if you integrate this provider.

UI kits for AI apps

• Gradio: [email protected] — 2026-10-08
• Features #13893 0d59e5e - Add an app view to gr.Workflow. Thanks @abidlabs ! #13866 ef1a961 - gr.Workflow: "Run this node" now reuses up-to-date upstream results instead of re-executing the whole upstream subgraph (Shift+click forces a full re-run). Also keeps a node's complet...
• Product update — skim if you use this tool.
• Gradio: @gradio/[email protected] — 2026-10-08
• Fixes #13936 0a7c8da - Fix hidden Accordion revealed with open=True rendering empty. Thanks @abidlabs !
• Product update — skim if you use this tool.
• Gradio: @gradio/[email protected] — 2026-10-08
• Fixes #13936 0a7c8da - Fix hidden Accordion revealed with open=True rendering empty. Thanks @abidlabs !
• Product update — skim if you use this tool.

Vector DBs

• Weaviate: Weaviate security release - Medium severity fix for a cross-collection vector disclosure in nearObject (CVE pending assignment) — 2026-10-15
• Weaviate v1.39.3 fixes a medium-severity authorization issue in the nearObject GraphQL operator.
• Security-related — prioritize review if you run this in production.
• Pinecone: Product Oct 15, 2026 On-demand mode for Pinecone BYOC Manish Talreja — 2026-10-15
• Product update — skim if you use this tool.
• Pinecone: Engineering, Research Sep 17, 2026 VQ-bench: a Composable Vector Quantization Framework Ashwin , Amir , Edo — 2026-10-15
• Product update — skim if you use this tool.
• Pinecone: Product Sep 9, 2026 Full-Text Search is Now Generally Available In Pinecone Database Manish , Gavin — 2026-10-15
• Product update — skim if you use this tool.
• Weaviate: PathSeer: Adaptive Filtered Vector Search — 2026-10-14
• Weaviate 1.40 adds PathSeer, a third HNSW filter strategy from a SIGMOD 2026 paper, contributed by the paper's first author. How it works, and how it compares to sweeping and ACORN.
• Product update — skim if you use this tool.

Source notes

1 source(s) failed this run:

• `stability` — HTTPConnectionPool(host='stability.ai', port=80): Max retries exceeded with url: /news-updates (Caused by ConnectTimeoutError( , 'Connection to stability.ai timed out. (connect timeout=20)'))

---

You're reading a draft of Model Drop. Swap in Beehiiv when ready — see README.