BeamWeaver

v0.1.10 Feature

This release adds 2 notable features for engineering teams evaluating rollout.

Published 17d AI Agents & Assistants

View tool

✓ No known CVEs patched

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

✓ No known CVEs patched in this version

Topics

ai-agents deepagents elixir harness human-in-the-loop langchain

+4 more

langgraph-alternative llm observability distributed-tracing

Summary

AI summary

Added new chat-model streaming APIs and improved handling of xAI Grok reasoning streams.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Adds `BeamWeaver.Core.ChatModel.stream_typed_events/3` and `async_stream_typed_events/3` functions. Adds `BeamWeaver.Core.ChatModel.stream_typed_events/3` and `async_stream_typed_events/3` functions. Source: llm_adapter@2026-07-16 Confidence: high	—
Feature	Medium	Sets xAI `grok-4.5` as default Grok chat profile with 500k context and multimodal capabilities. Sets xAI `grok-4.5` as default Grok chat profile with 500k context and multimodal capabilities. Source: llm_adapter@2026-07-16 Confidence: low	—
Feature	Medium	Agent model nodes use typed request‑level streaming when `stream: true` and provider supports typed events. Agent model nodes use typed request‑level streaming when `stream: true` and provider supports typed events. Source: llm_adapter@2026-07-16 Confidence: low	—
Feature	Low	Sets xAI `grok-4.5` as the default Grok chat profile with 500k context, multimodal input capability metadata, configurable reasoning metadata, current token pricing, and documented aliases. Sets xAI `grok-4.5` as the default Grok chat profile with 500k context, multimodal input capability metadata, configurable reasoning metadata, current token pricing, and documented aliases. Source: granite4.1:30b@2026-07-16-audit Confidence: low	—
Feature	Low	Agent model nodes use typed request-level streaming when `stream: true` is present and the provider supports typed events, forwarding live model reasoning chunks through graph and agent streams while reconstructing the final assistant message from visible answer/tool chunks. Agent model nodes use typed request-level streaming when `stream: true` is present and the provider supports typed events, forwarding live model reasoning chunks through graph and agent streams while reconstructing the final assistant message from visible answer/tool chunks. Source: granite4.1:30b@2026-07-16-audit Confidence: low	—
Bugfix
Bugfix	Medium	OpenAI‑compatible Responses streaming now recognizes xAI `response.reasoning_text.delta` and `response.reasoning_text.done` events. OpenAI‑compatible Responses streaming now recognizes xAI `response.reasoning_text.delta` and `response.reasoning_text.done` events. Source: llm_adapter@2026-07-16 Confidence: high	—
Bugfix	Medium	Reasoning stream chunks emitted as `%BeamWeaver.Core.ContentBlock.Reasoning{}` instead of generic maps, preventing folding into assistant answer text. Reasoning stream chunks emitted as `%BeamWeaver.Core.ContentBlock.Reasoning{}` instead of generic maps, preventing folding into assistant answer text. Source: llm_adapter@2026-07-16 Confidence: high	—
Bugfix	Medium	Agent streaming falls back to `stream_response/3` path when provider reports typed stream events as unsupported. Agent streaming falls back to `stream_response/3` path when provider reports typed stream events as unsupported. Source: llm_adapter@2026-07-16 Confidence: high	—

Full changelog

0.1.10 - 2026-07-08

Added

Added BeamWeaver.Core.ChatModel.stream_typed_events/3 and
async_stream_typed_events/3 for standalone chat-model callers that need
normalized BeamWeaver stream envelopes without relying on provider-specific
lifecycle event APIs.

Changed

Added xAI grok-4.5 as the default Grok chat profile with 500k context,
multimodal input capability metadata, configurable reasoning metadata, current
token pricing, and documented latest aliases.
Agent model nodes now use typed request-level streaming when stream: true
is present and the provider supports typed events, forwarding live model
reasoning chunks through graph and agent streams while keeping the final
assistant message reconstructed from visible answer/tool chunks.

Fixed

OpenAI-compatible Responses streaming now recognizes xAI
response.reasoning_text.delta and response.reasoning_text.done events
alongside OpenAI reasoning-summary events, so Grok reasoning can surface as
live %BeamWeaver.Stream.Events.MessageChunk{} reasoning blocks.
Reasoning stream chunks are now emitted as
%BeamWeaver.Core.ContentBlock.Reasoning{} instead of generic text-bearing
maps, preventing reasoning deltas from being folded into assistant answer
text by message projection or streamed final-response reconstruction.
Agent streaming falls back to the existing stream_response/3 path when a
provider reports typed stream events as unsupported, preserving existing
streaming behavior for providers that have not added typed event support.

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 BeamWeaver

Get notified when new releases ship.

About BeamWeaver

All releases →