# Ollama v0.6.8 — Release & Upgrade Intelligence
- **Product:** Ollama (https://ollama.com)
- **Version:** v0.6.8
- **Released:** May 03, 2025
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/ollama/updates/v0.6.8/

## Executive Summary (TL;DR)
Ollama v0.6.8. Source highlight: Performance improvements for Qwen 3 MoE models (`30b-a3b` and `235b-a22b`) on NVIDIA and AMD GPUs

## Improvements
- Performance improvements for Qwen 3 MoE models (`30b-a3b` and `235b-a22b`) on NVIDIA and AMD GPUs
- Reduced out of memory errors by improving worst-case memory estimations

## Bug Fixes
- Fixed `GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed` issue caused by conflicting installations
- Fixed a memory leak that occurred when providing images as input
- Fix issue that resulted in a `context canceled` error