v1.7.1
gohugoio/hugov1.7.1Sep 3, 2026by github-actions[bot]
Upgrading?
AI Summary
Focuses on tracking token usage and optimizing performance, introducing granular pricing models and improved subagent usage attribution.
Key Highlights
- Enhanced token usage tracking with speed-keyed buckets and repricing fields.
- Improved subagent usage attribution to their own sessions.
- Kept model context tiers and fast multipliers in the pricing catalog.
- Performance optimization by holding sweep duty cycles across operations.
New Features
- Token usage sweep discovery.
- Request speed and Codex service tier extraction.
- Price entries with long-context tiers.
- Speed-keyed buckets for usage tracking.
- Subagent usage attribution.
Full Release Notes
## git-ai v1.7.1 ### Release Channel `latest` > **Beta:** The Windows MSI and macOS PKG installers are beta. --- > **Detailed AI-generated changelog is being generated and will appear here shortly...** --- ### Installation Install this specific version with checksum verification: **macOS/Linux:** ```bash curl -fsSL https://github.com/git-ai-project/git-ai/releases/download/v1.7.1/install.sh | bash ``` **Windows (PowerShell):** ```powershell irm https://github.com/git-ai-project/git-ai/releases/download/v1.7.1/install.ps1 | iex ``` ### Downloads - **Linux (x64)**: `git-ai-linux-x64` - **Linux (ARM64)**: `git-ai-linux-arm64` - **Windows (x64)**: `git-ai-windows-x64.exe` - **Windows (ARM64)**: `git-ai-windows-arm64.exe` - **Windows MSI (x64)**: `git-ai-windows-x64.msi` - **Windows MSI (ARM64)**: `git-ai-windows-arm64.msi` - **macOS (Intel)**: `git-ai-macos-x64` - **macOS (Apple Silicon)**: `git-ai-macos-arm64` - **macOS PKG (Intel)**: `git-ai-macos-x64.pkg` - **macOS PKG (Apple Silicon)**: `git-ai-macos-arm64.pkg` - **macOS PKG (Universal)**: `git-ai-macos-universal.pkg` ### SHA256 Checksums ``` $(cat release/SHA256SUMS) ``` ### Attestation Verification Verify build provenance with the GitHub CLI: ```bash gh attestation verify git-ai-<platform> --repo git-ai-project/git-ai ``` <!-- Release notes generated using configuration in .github/release.yml at main --> ## What's Changed ### Other Changes * Log token usage sweep discovery by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2229 * fix(daemon): never run family side effects inline on the trace ingest worker by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2255 * feat(daemon): log checkpoint admission delayed behind trace ingestion by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2256 * feat(pricing): keep models.dev context tiers and fast multipliers in the catalog by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2232 * feat(token-usage): extract request speed / Codex service tier per entry by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2233 * feat(token-usage): price entries with long-context tiers and fast multipliers by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2234 * feat(token-usage): speed-keyed buckets and repricing fields on the wire by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2235 * feat(token-usage): match forked Codex sessions against their parent's usage prefix by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2236 * feat(token-usage): attribute subagent usage to the subagent's own session by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2241 * perf(token-usage): hold the sweep duty cycle across small files, emission, and reconcile by @svarlamov-git-ai in https://github.com/git-ai-project/git-ai/pull/2254 **Full Changelog**: https://github.com/git-ai-project/git-ai/compare/v1.7.0...v1.7.1
Related Repositories
ollama/ollama★ 184,215
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
fatedier/frp★ 110,053
A fast reverse proxy to help you expose a local server behind a NAT or firewall to the internet.
infiniflow/ragflow★ 92,614
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
junegunn/fzf★ 83,715
:cherry_blossom: A command-line fuzzy finder
caddyserver/caddy★ 78,239
Fast and extensible multi-platform HTTP/1-2-3 web server with automatic HTTPS