LeemerChat Release Timeline
Every release, every highlight
Track what shipped, when it shipped, and how each version evolved the LeemerChat intelligence workspace.
Latest release spotlight
Frontier models are now cleaner, stronger, and easier to pick.
v0.7.2 moves MiMo V2 Pro, GLM-5V Turbo, and Gemma 4 31B IT into the free partner tier, while GPT-5.4, GLM-5, and MiniMax M2.7 now anchor the premium stack.
Free partner models
3
MiMo V2 Pro, GLM-5V Turbo, and Gemma 4 now ship in the free tier.
Premium frontier
3
GPT-5.4, GLM-5, and MiniMax M2.7 now define the paid lineup.
Selector cleanup
Fewer overlaps
Older GPT chat variants, GLM-4.7, and the MiniMax split are gone from active runtime configs.
Latest
v0.7.2
Frontier Model Lineup Refresh
April 4, 2026
- • Added MiMo V2 Pro, GLM-5V Turbo, and Gemma 4 31B IT as free partner models.
- • Promoted GPT-5.4, GLM-5, and MiniMax M2.7 across premium model surfaces and agent configs.
- • Removed older GPT-5 chat variants plus the GLM-4.7 and MiniMax M2.1/M2.5 split from active runtime configs.
- • Tips popup now points to the new frontier launch story and the live localized $1 first-month offer.
v7.0
v7 Model Refresh
March 18, 2026
- • Grok 4.20 Beta replaces Grok 4 with a faster flagship profile and a 2M context window.
- • GPT-5.4 Mini and GPT-5.4 Nano replace earlier small GPT variants for lower-latency workloads.
- • Mistral Small 4 consolidates reasoning, multimodal, and coding improvements into a single lighter-tier option.
- • NVIDIA Nemotron 3 Super 120B joins the free partner lineup with a 1M context window.
- • Deep Research pipeline picks up the latest model lineup across supporting agents.
v0.7.1
Deep Research Frontier v2 Agentic Rewrite
March 6, 2026
- • Lead frontier now drives evidence-led discovery across Firecrawl + Exa tool flows.
- • Sub-agents run iterative refinement rounds with follow-up prompts and targeted deep scraping.
- • Evidence ledger and reconciliation milestones are persisted for transparent research transparency.
- • Deep-research UI now surfaces live evidence activity, open questions, and claim-state.
v0.6.3
Token Accuracy & Output Cleanup
February 24, 2026
- • Token counter now includes full conversation context plus current draft input.
- • Automatic cleanup path for non-markdown assistant output before final display/save.
- • Improved readability and context-window accuracy in day-to-day chat usage.
v0.6.1
Gradient Revamp & Robustness
February 23, 2026
- • Completely redesigned all theme gradients with cleaner 3-stop 135deg flows.
- • Frontend snappiness fixes for the model selector dropdown.
- • Prompt generation JSON parsing robustness and bug fixes.
v0.6.0
KingLeemer Beta & Multi-Model Chat
February 2026
- • KingLeemer V1, V2, and V3 autonomous agents available via Beta Mode.
- • Multi-Model Collaboration with @mention chip system.
- • Deep Research (/research) page with multi-stage pipeline and email delivery.
- • Model Usage Analytics, Currency/Crypto tools, and Daily Prompts Cache.
v5.3.0
Major UI Modernization
February 2026
- • Modern CSS utilities with animations and glass surface effects.
- • Redesigned welcome screen, chatbox, sidebar, and model selector.
- • Fixed font system with next/font/google for better compatibility.
- • Cleaner message bubbles and refined visual hierarchy throughout.
v5.2.5
Global Agent Integrations
February 2026
- • Blackbox.ai agents now launch directly from LeemerChat.
- • Cursor Cloud agents can be dispatched from chat for repo workflows.
- • Bring your API keys and spend existing credits inside LeemerChat.
v5.2.4
Visual Refresh & SEO
February 2026
- • Geist font family (Sans, Mono, Pixel) for modern typography.
- • Flow field particle background with mouse interaction on landing.
- • SEO improvements with X-Robots-Tag headers for proper indexing.
- • Token counter, draft auto-save, and enhanced attachment previews.
v5.2.3
PDF Jobs Manager
January 2026
- • Job history dashboard for every processed PDF.
- • Status tracking, retry, and download management in one view.
- • Built on the revamped PDF architecture from v5.2.2.
v5.2.2
Revamped PDF Architecture
January 2026
- • 85% cost reduction with a single-model pipeline.
- • 70% faster processing and unified question detection + solving.
- • Native PDF parsing and smart fallbacks for reliability.
v5.1.2
PDF AI Processing
January 2026
- • Automatic question detection and solving for PDFs.
- • Professional answer sheet generation with math formatting.
- • Multi-agent consensus for higher reliability.
v5.1.1
Speed + Observability
January 2026
- • Time-to-first-token reduced to ~2 seconds for fast models.
- • Instrumentation upgrades across the generation pipeline.
- • Streamlined worker processing for lower API overhead.
v5.1.0
Codebase Chat + AI Memory
January 2026
- • GitHub-connected Codebase Chat for repo-aware assistance.
- • AI Memory preferences synced across devices.
- • Second Thought expert consultations for verification.
v5.0
Auto-Research v5
December 2025
- • Simplified research engine powered by Perplexity Sonar Pro.
- • Unified report + podcast job history.
- • Mobile-first research UX refresh.
v4.9
PowerCode + Model Refresh
Late 2025
- • PowerCode coding agent with sandbox execution.
- • LeemerGLM-106B refresh and new model lineup.
- • Dynamic contextual loading phrases.
v4.8
Smoother Experience
December 2025
- • IKEA-inspired frosted glass interface.
- • Durable generation for better chat reliability.
- • Performance and latency improvements across chat.