Llmff

v1.1.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 1mo LLM Frameworks

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agent-tools cli inference-pipeline json-schema llm llm-inference

+13 more

llm-pipeline llmops local-ai local-llm observability offline ollama openai-compatible pipeline reproducible retrieval rust yaml

Summary

AI summary

Adds a bounded sequential op: loop stage for retry-style LLM workflows.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Adds bounded sequential `op: loop` stage for refinement and retry workflows. Adds bounded sequential `op: loop` stage for refinement and retry workflows. Source: llm_adapter@2026-06-08 Confidence: high	—
Feature	Medium	Provides loop output metadata: `iterations_run`, `stop_reason`, and `final_stage`. Provides loop output metadata: `iterations_run`, `stop_reason`, and `final_stage`. Source: llm_adapter@2026-06-08 Confidence: high	—
Feature	Medium	Adds trace/event loop context fields: `loop_id`, `loop_iteration`, and `loop_stage_id`. Adds trace/event loop context fields: `loop_id`, `loop_iteration`, and `loop_stage_id`. Source: llm_adapter@2026-06-08 Confidence: high	—
Feature	Medium	Enables `inspect --format json` to show loop metadata including body size and maximum expanded stage count. Enables `inspect --format json` to show loop metadata including body size and maximum expanded stage count. Source: llm_adapter@2026-06-08 Confidence: high	—
Feature	Medium	Ensures compatibility with Python subprocess supervisor, Node.js streaming supervisor, agent runner guide, and OpenTelemetry bridge. Ensures compatibility with Python subprocess supervisor, Node.js streaming supervisor, agent runner guide, and OpenTelemetry bridge. Source: llm_adapter@2026-06-08 Confidence: high	—

Full changelog

llmff v1.1.0

llmff v1.1.0 adds a bounded sequential op: loop stage for inspectable
refinement and retry-style LLM workflows.

Added

op: loop with required max_iterations, explicit break_on, and embedded
body stages.
Loop output metadata with iterations_run, stop_reason, and final_stage.
Trace and event loop context fields: loop_id, loop_iteration, and
loop_stage_id.
inspect --format json loop metadata showing body size and maximum expanded
stage count.
Compatibility with the existing Python subprocess supervisor, batch supervisor,
Node.js streaming supervisor, agent runner adoption guide, OpenTelemetry bridge,
and ecosystem readiness gates.

Boundary

Loops are bounded execution stages. They are not a scheduler, memory system,
policy engine, or autonomous planner.

Install

cargo install --git https://github.com/syndicalt/llmff --tag v1.1.0 llmff
scripts/smoke-install.sh --git https://github.com/syndicalt/llmff --tag v1.1.0

Release preflight

Run the release preflight gate before creating or pushing the tag:

scripts/release-preflight.sh v1.1.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track Llmff

Get notified when new releases ship.

About Llmff

All releases →