ollama · Issues· 4033 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #18394
Include the served manifest digest in local /api/chat responses (tested v0.34.0 patch)
Updated Sep 17, 2026 - #18505
[BUG] MLX nvfp4: admitted request stalls in prefill at processed=total-1 with zero tokens for minutes under sustained single-slot load; only runner SIGTERM recovers
Updated Sep 17, 2026 - #18509
Ollama refusing tool role, which always worked fine in llama.cpp with qwen.
bugUpdated Sep 17, 2026 - #18507
Windows 11 26200, Ollama 0.34.1 — tray app shows icon but never starts server; manual ollama serve works perfectly
bugUpdated Sep 17, 2026 - #18506
glm-5.3-flash intermittently emits malformed string-encoded tool calls via Ollama Cloud
Updated Sep 17, 2026 - #18490
Feature Request: Restore built-in agent as an opt-in CLI command / launcher option
feature requestUpdated Sep 17, 2026 - #16892
glm-ocr infinite loop
bugUpdated Sep 17, 2026 - #18502
linux: fix Vulkan inference support on ARM64 (PR implemented)
bugUpdated Sep 17, 2026 - #18494
qwen3-vl:8b-instruct 0xc0000005 on Vulkan AMD RX 6750 XT after multi-model load (Windows 0.34.1)
Updated Sep 16, 2026 - #18396
Jetson Orin Nano 8GB: Gemma 4 E4B OOM with --load-mode dio, while identical configuration succeeds without DIO
bugUpdated Sep 16, 2026 - #17638
gpt-oss: HTTP 500 "error parsing tool call" — ollama rejects a tool call its own model generated
bugUpdated Sep 16, 2026 - #18491
Connect to Claude Desktop forks into a new anonymous account instead of the signed-in session, and Disconnect intermittently fails to restore it
Updated Sep 16, 2026 - #15142
Add Mistral Small 4 to Ollama Models
modelUpdated Sep 16, 2026 - #18487
Feature request: optional external resource lock for shared GPU coordination
feature requestUpdated Sep 16, 2026 - #18483
minicpm5-2b native tool calls never parse
bugUpdated Sep 16, 2026