Skip to content

Reame

v0.1.5 Feature

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

Published 13d Model Serving & MLOps
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

cpu gguf inference kv-cache llama-cpp llm
+4 more
local-llm openai-api self-hosted speculative-decoding

Summary

AI summary

Reame now automatically downgrades decoding for Qwen3.5 models to prevent crashes and adds Qwen3.5-9B as the judgment‑model pick with a one‑liner SEO audit command.

Changes in this release

Feature Low

Qwen3.5-9B added to the model catalog as the judgment pick for SEO audits.

Qwen3.5-9B added to the model catalog as the judgment pick for SEO audits.

Source: llm_adapter@2026-07-15

Confidence: high

Feature Low

README includes a verified one-liner to audit any live page from its raw HTML using Qwen3.5-9B.

README includes a verified one-liner to audit any live page from its raw HTML using Qwen3.5-9B.

Source: llm_adapter@2026-07-15

Confidence: high

Dependency Low

Prebuilt binaries now support Linux x64, Linux ARM64, and macOS ARM64 with libzstd runtime dependency.

Prebuilt binaries now support Linux x64, Linux ARM64, and macOS ARM64 with libzstd runtime dependency.

Source: llm_adapter@2026-07-15

Confidence: high

Bugfix Medium

Reame no longer crashes on recurrent and hybrid models during generation.

Reame no longer crashes on recurrent and hybrid models during generation.

Source: llm_adapter@2026-07-15

Confidence: high

Full changelog

v0.1.5 — the judgment release

Recurrent and hybrid models no longer crash. Qwen3.5's delta-net layers (like Mamba and Jamba states) cannot roll back past positions, which speculative decoding's rejection step requires — so reame run died mid-generation with a cryptic "failed to truncate sequence" on the whole Qwen3.5 family. Reame now detects these architectures and downgrades to classic decoding automatically, logging why.

Qwen3.5-9B joins the catalog as the judgment pick. We measured four models on a real task — an SEO audit of a live webpage: a 1.5B and OLMoE invented findings (flagged missing alt text that exists), a 30B-A3B miscounted the meta description length, and the 9B got everything right, including an empty <h2> the page really has, in 73s on a laptop (16.6 tok/s plain decode, 5.3 GB). reame run qwen3.5-9b and you have it. The README gains a verified one-liner that audits any live page from its raw HTML.

Extraction stays the small models' turf — that's the measured boundary: when the answer lives in the context, run a 1.5B at 81 tok/s; when the task needs judgment, batch the 9B.

Prebuilt binaries: Linux x64, Linux ARM64, macOS ARM64 (runtime dependency: libzstd).

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

Track Reame

Get notified when new releases ship.

Sign up free

About Reame

All releases →

Related context

Related tools

Beta — feedback welcome: [email protected]