Ollama

Track Ollama development. Run large language models locally.

About this developer podcast

Ollama turns GitHub activity into short audio updates.

Episodes cover commits, pull requests, issues, and release work.

Use this page to listen, subscribe, and find related developer podcasts.

This show page is the public hub for the podcast. It has one stable listen URL.

Use it to find RSS feeds, recent episodes, related shows, and older episode links.

If the project is quiet, new episodes may pause until there is useful activity to summarize.

Recent episodes from Ollama

  1. Native Decision Models and Precision Fixes

    Ollama shifted decision-model scoring into upstream llama.cpp and added multimodal embedding support, while fixing tool-call precision and model loading reliability. The pattern is less custom code, fewer edge-case failures.

  2. Weekly Recap - Decision Models Go Native, Embeddings Go Multimodal

    Ollama shifted decision-model scoring into upstream llama.cpp and added multimodal embeddings, alongside fixes for OpenAI compatibility and Windows model storage. The pattern this week is retiring custom patches in favor of native runner…

  3. GPU Discovery Fallback and Docs Cleanup

    Twelve pull requests were dominated by ten low-risk docs fixes, with the functional impact concentrated in PR #18923 adding a native-probe fallback for empty GPU discovery output, plus a sidebar load-animation fix in PR #18937.

  4. OpenAI Parity and Reasoning Fixes

    Ollama activity centers on OpenAI compatibility and model correctness, with streaming and tool-history fixes, Pi thinking controls and parser repairs, plus a merged change to skip a costly model migration.

  5. Legacy Model Cleanup and Agent Launcher Polish

    Ollama is retiring legacy model compatibility shims in favor of on-load conversion, while tightening its launch experience with Claude Code and Codex. Inference fixes for reasoning tags, tool parsing, and raw output round out the activity.

  6. Memory Accuracy, Tool Fixes, and Simpler Onboarding

    The big pattern is correctness under real-world conditions — VRAM accounting, tool-call rendering, and Windows and cross-platform edge cases. Onboarding gets simpler while server debugging gets more observable.

  7. Multimodal Embeddings and Cloud Convergence

    Ollama added multimodal embeddings and tightened local-cloud integration, while landing reliability fixes for Windows, OpenAI compatibility, and model listing. The pattern is expansion of the embed API paired with smoothing the cloud…

  8. MLX Performance Push and Tool-Call Reliability

    The dominant pattern is MLX work to cut latency and memory overhead on Mac GPUs, backed by runtime version bumps. A second cluster tightens OpenAI compatibility and model parser handling for tool calls, plus reliability fixes for…

  9. Tokenizer Parity and RC-Aware Updates
  10. Weekly Recap - System One Decisions Take Center Stage
  11. MLX Tokenizer Fixes and Thinking Model Reliability
  12. Tool Call Ordering Fixes and Decision Models
  13. Decision Models Take Shape
  14. Pointer-Head Scoring and Reliability Fixes
  15. System One Moves to Explicit Capabilities
  16. System One Decisions Launch
  17. Silent Sampling Overrides Fixed
  18. Weekly Recap - Single-Pass Structured Outputs and MLX Speedups
  19. Tool-Call Reliability Fixes
  20. Desktop Fixes and API Correctness
  21. Mac Polish, MLX Reliability, and Compatibility Fixes
  22. GPU Efficiency Fix and MLX Hardening
  23. Single-Pass Structured Outputs and MLX Speedups
  24. Agent Reliability Fixes and Offline Model Transfer
  25. Thinking Separation Fix and GPU Visibility
  26. Weekly Recap - MLX Graduation, Reasoning Fixes, and Onboarding Polish
  27. Compatibility Fixes and Mac Efficiency
  28. Closing the Gaps Between Model Metadata and Client Compatibility
  29. Hardening Pass Across Network, Memory, and Release Paths
  30. The MLX Engine Comes Home
  31. Streaming Reliability and Parser Cleanup
  32. MLX Comes of Age, Cancellation Gets Fixed
  33. Consistency Fixes for Tool Schemas and Image Input
  34. Weekly Recap - Memory Discipline and Tool-Calling Cleanup
  35. Precision Fixes and Cleanup
  36. Silent Failures Get Loud
  37. Taming Memory in the MLX Runner
  38. Tool-Call Parsing Gets a Hardening Pass
  39. Codex Integration Hardening and Streaming Reliability
  40. Tightening Up the OpenAI Compatibility Layer
  41. Closing the Gaps Between What You Configure and What Actually Runs
  42. Weekly Recap - MLX Hardens as a First-Class Engine
  43. Context Handling Gets Serious
  44. Codex Integration Push and Schema Hardening
  45. Structured Output Meets Speculative Decoding
  46. Hardening the MLX Engine and Fixing What Feeds the Scheduler
  47. MLX Grows Up, Model Intent Gets Respected
  48. Runner Reliability and a Documentation Cleanup Wave
  49. Proxy Fix and a Documentation Cleanup Sweep
  50. Weekly Recap - Claude Desktop Integration & MLX Engine Maturity
  51. Correctness Fixes Under the Hood
  52. Agent Expansion, Desktop Polish, and a Documentation Sweep
  53. MLX Engine Grows Up
  54. Desktop App Stability and the Claude Integration Cleanup
  55. MLX Runner Hardens Up, Claude Desktop Gets Real Estate
  56. Claude Desktop Integration Overhaul
  57. Hardening the Runner Scheduler
  58. Weekly Recap - Desktop Apps Grow Up, Inference Gets More Reliable
  59. Visibility Into What the Model Actually Sees
  60. Fixing the Long-Prompt Timeout Trap
  61. Claude Desktop Gets Deep App Integration
  62. Closing the Gaps on Hung Requests and Broken Installs
  63. Closing the Gap on Model Metadata Overhead
  64. Silent Failures Get Loud
  65. Fixing the Small Gaps Between Local and Cloud
  66. Weekly Recap - Model Compatibility and the Push Into Coding Agents
  67. Silent Failures Get Louder
  68. Qwen Gets Serious About Coding Agents
  69. Launch Ecosystem Expands, Parser Edge Cases Get Squashed
  70. Backup Collisions and a Security Fix Converge
  71. Fixing the Defaults That Silently Hurt You
  72. Model Coverage and Tool-Call Reliability
  73. Vision Support Lands in the MLX Runner
  74. Weekly Recap - Vision Runners and Robustness Cleanup
  75. Edge Case Cleanup Across the Stack
  76. Speculative Decoding Gets a New Draft Architecture
  77. Crash Fixes and the Launcher Ecosystem Grows
  78. Silent Data Corruption Fixes
  79. Hardening the Thinking and Tool-Call Pipeline
  80. Closing the Gaps Where Failures Go Silent
  81. Streams That Fail Silently, Fixed
  82. Weekly Recap - Speed, Speculation, and Scheduler Reliability
  83. Scheduler Reliability Overhaul
  84. Speculative Decoding and the Cloud Model Nudge
  85. API Compatibility and Model Correctness Push
  86. Speculative Decoding Gains and a Lint Lockdown
  87. Cleaning Up Concurrency and Cutting Experimental Code
  88. Trust Your Cache, Fix Your Logs
  89. Parsing Bugs and Process Hardening
  90. Weekly Recap - New Model Support & Concurrency Hardening
  91. Closing the Gaps Between Local Config and Cloud Compatibility
  92. MLX Performance Push and a Scheduler Race Fix
  93. A Server Reliability Sweep
  94. Tool Calling Gets a Reliability Pass
  95. Scheduler Race Conditions Take Center Stage
  96. Command Line Consolidation and Hardware Fixes
  97. Thinking Models Leak Control Tokens
  98. Weekly Recap - Hardening the Runtime, One Edge Case at a Time
  99. Closing the Gaps Between API Layers
  100. Cleaning Up the Agent Loop and Widening Model Support

Episode archive for Ollama

  1. Agent Package Gets a Major Consolidation
  2. Agent Hardening and a Path Traversal Fix
  3. Resource Leaks and GPU Placement Get a Clean-Up Pass
  4. Cleaning Up Memory Planning and the Launch Experience
  5. Fixing What Defaults Got Wrong
  6. Weekly Recap - Model Correctness and Runtime Hardening
  7. Tightening Up the Serving Layer
  8. Thinking Model Output Is Getting a Real Fix
  9. Stability Sweep Across Cloud, GPU, and Agent Tools
  10. Qwen3.5 Gets Untangled, and Small Bugs With Big Blast Radius
  11. Model Behavior Bugs and Blob Security Hardening
  12. Launch Gets Smarter, Model Support Gets Wider
  13. Cleaning Up the Scheduler's Edge Cases
  14. Weekly Recap - Rebuilding the Model Pipeline and Tightening the Guardrails
  15. Truth in Reporting
  16. MLX Create Pipeline Rewrite Lands
  17. Agent Harness Lands, Hardware Support Gets a Cleanup
  18. Gemma 4 Support and Platform Improvements
  19. Weekly Recap - MLX Performance & Path Handling
  20. Memory Management and Multimodal Parsing Fixes
  21. GPU Offloading and Tool Call Fixes
  22. Performance Optimizations and Model Handling Improvements
  23. Infrastructure Updates and Platform Fixes
  24. Multimodal Fixes and Developer Experience Updates
  25. Cache Architecture Overhaul and Data Race Fixes
  26. Developer Tools and Cross-Platform Reliability
  27. Weekly Recap - Integration Expansion & Server Reliability
  28. Audio Support and Infrastructure Refinements
  29. Integration Ecosystem and API Consistency Push
  30. Platform Integration Expansion and API Reliability Fixes
  31. Model Integration and Windows System Improvements
  32. LLaMA Server Integration Hardening
  33. Integration Platform Expansion
  34. Model Integration Updates
  35. Weekly Recap - Infrastructure Modernization
  36. Major Architecture Overhaul Removes CGO Dependencies
  37. MLX Model Display Fixes and Template Parser Cleanup
  38. Weekly Recap - Performance Optimization & Launch System Improvements
  39. DFlash Speculative Decoding Rollback
  40. Model Inventory Refactoring
  41. Startup Performance Optimization
  42. Codex Integration Enhancement
  43. Weekly Recap - MLX Performance & Codex Integration
  44. Release Build Optimization
  45. Speculative Decoding and Codex App Updates
  46. MLX Sampler Overhaul and Codex Integration
  47. Vision Model Integration Enhancement
  48. MLX Threading and Claude Image Fixes
  49. Model Transfer Optimization and Test Reliability
  50. Claude Desktop Integration Removed
  51. Launch Command Enhancements
  52. Speed Revolution - MTP Decoding and Smart Caching
  53. Go 1.26 Runtime Update
  54. Weekly Recap - MLX Threading & Model Recommendations
  55. MLX Threading Fixes and Claude App Integration
  56. Model Recommendations and Windows Gateway Fix
  57. Metal GPU Stability and Gemma4 Updates
  58. Launch Experience Improvements and Model Recommendations
  59. Multi-Sequence Batching and New Model Support
  60. Tokenizer Bug Fix for BPE Processing
  61. Weekly Recap - MLX Performance & Launch Integrations
  62. MLX Sampling Performance Enhancement
  63. OpenAI Reasoning Integration
  64. Launch System Improvements and Integration Fixes
  65. Launch System Overhaul and Documentation Updates
  66. MLX Performance Boost and Model Updates
  67. New CLI Integration and Performance Improvements
  68. Weekly Recap - MLX Performance & Launch Integration Expansion
  69. MLX Sampler Improvements
  70. Windows WSL Integration Simplified
  71. Gemma4 Enhancements and Copilot CLI Integration
  72. Hermes Agent Integration and Gemma4 Improvements
  73. Gemma 4 MLX Support and Mixed-Precision Improvements
  74. Weekly Recap - Model Integration and Tooling Enhancements
  75. ROCm 7.2.1 Performance Update
  76. Gemma4 Parser Improvements
  77. Model Updates and Tool Call Fixes
  78. Error Handling and Modelfile Fixes
  79. Weekly Recap - Gemma4 Integration & Audio Support
  80. Performance Lessons and Gemma4 Refinements
  81. Gemma4 Arrives with Audio Magic
  82. Modernizing Codex Configuration
  83. Tokenizer Love and Better Model Support
  84. Legacy Compatibility and Developer Experience Wins
  85. Smoothing the Launch Experience
  86. Fixing the Inconsistencies That Matter
  87. Smart Caching and Better User Experience
  88. VS Code Integration Takes Center Stage
  89. Precision Revolution - New Float Formats and Testing Powerhouse
  90. MLX Performance Breakthrough and Smarter Caching
  91. Nvidia Partnership Takes Center Stage
  92. Bug Squashing Bonanza
  93. The Caching Revolution
  94. Bug Squashing and Launch Improvements
  95. Launch Command Gets a Major Polish
  96. Spring Cleaning and Performance Gains
  97. Thinking Streams and Local Tool Power-ups
  98. Stability First - Error Handling and Performance Fixes
  99. MLX Gets a Major Upgrade and Web Search Goes Live
  100. Simplifying the Sampling Story
  101. Cloud Models Get Smarter & Build Performance Boost
  102. Cloud Integrations Get Some Love
  103. Smarter Constraints and Qwen3.5 Boost
  104. Cloud Integration Drama and AI Model Expansion
  105. Smarter Sampling and Crash Prevention
  106. Building Bridges for Better Model Compatibility
  107. MLX Runner Gets Rock Solid
  108. Tool Calling Gets Smarter
  109. Cleaner Shutdowns and Faster Startups
  110. Qwen 3.5 Architecture Lands with Safety Upgrades
  111. Memory Management Revolution
  112. Nemotron Architecture Lands with Unified Cache Vision
  113. Fixing the WSL Plugin Problem
  114. Smarter UIs and Smoother Onboarding
  115. Tokenizer Consolidation & MLX Library Improvements
  116. Rolling Back and Rolling Forward
  117. Editor Integration Revolution
  118. MLX Display Bug Squashing Day
  119. MLX Runner Gets Major Model Upgrades
  120. MLX Performance Breakthrough and Anthropic Search
  121. MLX Runner Revolution and Documentation Polish
  122. Refactoring Rollercoaster and Developer Experience Wins
  123. Bug Squashing Bonanza
  124. Smooth Onboarding for New Users
  125. Polish and Perfectionism - The Art of Getting the Details Right
  126. Cleaning Up the Config Game
  127. Speed Boost and Model Magic
  128. Memory Magic and Command Makeover
  129. Making Ollama Play Nice with Everyone
  130. The Great Cleanup - Manifests Get Their Own Home
  131. New Model Architecture and Image Generation Fixes
  132. New Model Support and Memory Management Wins
  133. FLUX.2 Image Generation Arrives
  134. Image Generation Goes Native and Parser Cleanup Magic
  135. Dynamic Loading and Experimental Models Take Center Stage
  136. Release Day Rescue Mission

More public developer podcasts

Browse other public Podlog show hubs with crawler-readable RSS and episode links.

  1. Agent Of Empires Daily 94 episodes
  2. Buzz Transcription 102 episodes
  3. Frigate NVR Updates 242 episodes
  4. Ruby Core Updates 250 episodes
  5. Openharness Daily 67 episodes
  6. Linux Kernel 231 episodes
  7. Homebrew 254 episodes
  8. TypeScript 143 episodes
  9. Ruby on Rails 247 episodes
  10. Redis 222 episodes
  11. Next.js 256 episodes
  12. Vue.js 194 episodes
  13. Rust 227 episodes
  14. Python 240 episodes
  15. Go 240 episodes
  16. VS Code 258 episodes
  17. TailwindCSS 150 episodes
  18. Node.js 252 episodes
  19. Django 244 episodes
  20. React Native 201 episodes
  21. PostgreSQL 243 episodes
  22. Kubernetes 253 episodes
  23. LangChain 238 episodes
  24. PyTorch 257 episodes
  25. Maestro Daily 93 episodes
  26. Videorc Daily 87 episodes
  27. Agora Next Updates 274 episodes
  28. Rails Daily 258 episodes
  29. Home Assistant Daily 263 episodes
  30. Linux Kernel Daily 237 episodes
  31. React Daily 154 episodes
  32. AtomVM Daily 21 episodes
  33. Next.js Daily 258 episodes
  34. Pi Mono 138 episodes
  35. OpenClaw 215 episodes
  36. RuView 170 episodes
  37. Godot Daily 203 episodes
  38. Jabref Daily 104 episodes
  39. Jabref Daily 105 episodes
  40. Next Js Daily 93 episodes
  41. Headroom Daily 102 episodes
  42. NativeScript iOS Daily 44 episodes
  43. Navidrome Daily 115 episodes
  44. BlocksBeyondTheStars Daily 83 episodes
  45. Shannon 104 episodes
  46. Rimworld ChickenBatteryCage Daily 1 episode
  47. Eleviewer Daily 1 episode
  48. iCloud Photos Downloader 36 episodes
  49. The Algorithm Daily 20 episodes
  50. Agora Next Daily 16 episodes
  51. Onlook Design Updates 34 episodes
  52. Energy Inc Ace Mate Learning Daily 2 episodes
  53. tiny-gpu Daily 20 episodes
  54. OpenAI Skills 52 episodes
  55. Crush Daily 1 episode
  56. ControlNet Daily 0 episodes
  57. FinceptTerminal Daily 0 episodes