Docling
Docling adds three more OCR model presets and keeps fixing format edge cases.
A side-by-side editorial comparison of Recall and Userpilot — release velocity, themes, recent moves, and the top alternatives to consider.
Recall gives its chat and MCP-connected apps exact-phrase search and whole-card reading.
The 2 November release changes how Recall retrieves rather than what it stores: chat, which used to search only by meaning, can now match exact words and read whole cards, and apps such as Claude, ChatGPT and Cursor get the same tools through the Recall MCP server. Users can now see and manage which apps are connected. Gmail saving works better and age-restricted YouTube videos can be saved. September's work was about intake: EPUB books, multi-image cards, OCR on social posts and 1,000-link CSV imports.
Userpilot ships MCP support, connecting product analytics directly into AI development and agent workflows.
Userpilot has launched an MCP server that exposes its product context—feature adoption rates, friction points, cohort behavior—directly to AI coding tools and agents. In the same week, session replay gained automatic friction and error detection, reducing the analysis burden from manual review. Together these two July 2026 releases push the product toward surfacing insights automatically rather than requiring analysts to go looking.
The 2 November release changes how Recall retrieves rather than what it stores: chat, which used to search only by meaning, can now match exact words and read whole cards, and apps such as Claude, ChatGPT and Cursor get the same tools through the Recall MCP server. Users can now see and manage which apps are connected. Gmail saving works better and age-restricted YouTube videos can be saved. September's work was about intake: EPUB books, multi-image cards, OCR on social posts and 1,000-link CSV imports.
After two months of adding new things to save, Recall is now working on getting answers back out of a large library, both in its own chat and in outside assistants. Treating MCP as a main access path, with controls over connected apps, suggests the library is meant to serve other AI tools as well as Recall's own interface.
Expect more MCP-facing controls, such as per-app permissions or write access, along with more retrieval fixes for large knowledge bases, which this release already names as a focus.
Userpilot has launched an MCP server that exposes its product context—feature adoption rates, friction points, cohort behavior—directly to AI coding tools and agents. In the same week, session replay gained automatic friction and error detection, reducing the analysis burden from manual review. Together these two July 2026 releases push the product toward surfacing insights automatically rather than requiring analysts to go looking.
The pattern across recent releases is toward intelligent automation: session replay that flags errors without manual review, multi-channel workflow orchestration from a single builder, and now MCP that feeds product data into AI agents. Userpilot is repositioning from 'add in-app guides' to 'orchestrate the full product-led growth loop'—with AI doing the interpretation layer.
Expect more AI-mediated analysis features: automated insight generation, AI-drafted in-app messages based on behavior patterns, and deeper CRM integrations. The MCP release positions Userpilot to compete on intelligence against Pendo and Gainsight rather than purely on tooling breadth.
Other ai-assistants products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either Recall or Userpilot.
Docling adds three more OCR model presets and keeps fixing format edge cases.
Hyperscience leans on explainability and third-party proof to win regulated buyers.
InvokeAI's v7 alphas keep adding model families while v6 gets maintenance and small Canvas gains.
LobeChat follows its proactive-bot release with approval, task and Linear fixes in v2.3.1.
DocsBot Week adds Operator, an agent that builds bots and fixes their sources.
Ollama v0.40.15 adds MiMo V2.6 after a v0.40.14 pass on Claude Code and OpenAI-API fixes.
See all Recall alternatives → · See all Userpilot alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. Recall is currently shipping more aggressively (velocity 5.0 vs 0.0), with 0 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Recall is currently shipping more aggressively (velocity 5.0 vs 0.0), with 0 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other ai-assistants products to evaluate alongside.
Top Recall alternatives in ai-assistants are ranked by recent ship velocity. Browse the "Recall alternatives" section above for the current picks, or visit /alternatives/getrecall for the full list with editorial commentary on each.
Top Userpilot alternatives in ai-assistants are ranked by recent ship velocity. Browse the "Userpilot alternatives" section above for the current picks, or visit /alternatives/userpilot for the full list with editorial commentary on each.