Pi 1.5.1 released
A feature release adding shell authentication commands, improved cache efficiency, message threading for Anthropic, richer diagrams, and classifier model support.
What changed in 1.5.1
- Log in and out from the shell using
pi auth loginandpi auth logoutwith options for API keys and device-code sign-in. - Prompt caches now stay warm between runs for 30 minutes when refresh costs justify it, reducing redundant cache rebuilds.
- Anthropic requests now send only new messages while the server continues conversations, falling back to full requests when context changes.
- Mermaid diagrams now include pie charts, custom styles, and clickable labels; diagrams appear in branch and compaction summaries.
- Call Microsoft-Decision-1 classifier through
models.classify()for typed decisions and probabilities via GitHub Copilot.
Check before you update
- Custom
GrepOperations.readFilecallbacks no longer receive context lines; use grep'sexecute()override for remote system searches. - OpenAI-compatible models no longer send empty
toolsarray when tools are disabled; setcompat.requiresToolsWithToolHistory: trueif your proxy requires it after tool history exists.
Summarised by AgentGid from Earendil Inc. (created by Mario Zechner)'s official release notes on GitHub, as reported there; read the original before relying on it. Price, status and release dates come from our daily data.
Pi at a glance
| What it is | Minimal, extensible open-source coding agent harness with interactive TUI, print/JSON, RPC and SDK modes, created by Mario Zechner and now maintained by Earendil Inc. It supports 15+ model providers through API keys or … |
|---|---|
| Latest version | 1.5.1 is the latest stable release we track. |
| Release pace | 5 stable releases in the last 30 days; 1.5.1 came 7 days after the previous one |
| Price | Free (OSS). Free and MIT-licensed; costs are those of the chosen model provider. Pi Durable (experimental) is also free per the 1.0 announcement. (checked Nov 5, 2026) |
| Downloads, last 7 days | 6.2M up 5.4% |
| Gid Score | 75 #5 in Terminal coding agents |