Summary
The informers gem (RubyGems: informers, by Andrew Kane) is a Ruby library for running Hugging Face transformer models locally via ONNX Runtime — no API keys or network calls required. It provides a concrete execution surface for embeddings generation, text classification, and other transformer inference pipelines, and is not instrumented by this SDK.
What is missing
No instrumentation exists for any informers execution surface. Key APIs that should be instrumented:
Embeddings
Informers.pipeline("embedding", model_name).(text) — runs a local ONNX embedding model and returns vector output. This is the primary execution method analogous to embeddings.create in hosted embedding APIs.- Batch embedding calls (passing an array of texts) for semantic search / RAG pipelines.
Other Inference Pipelines
Informers.pipeline("text-classification", ...), "feature-extraction", "zero-shot-classification", "reranking", etc. — each .call invocation runs local model inference and is a natural span boundary (model name, input, output, latency).
Weekly downloads
5.5K (as of 2026-07-23; https://clickgems.clickhouse.com/dashboard/informers) — ClickGems reports Last day / Last week / Last month / Total download counts sourced from Ruby Central's public RubyGems download dataset. RubyGems.org itself only exposes "Total downloads" and "downloads for this version," not a weekly figure, so ClickGems was used as the directly-reported weekly stat rather than deriving one.
Additional signals: v1.3.0 released April 15, 2026 (rubygems.org); Total downloads 2,515,045 (rubygems.org).
Braintrust docs status
not_found — Checked https://www.braintrust.dev/docs/integrations/sdk-integrations (lists only openai, anthropic, ruby-llm, ruby-openai for Ruby) and searched https://www.braintrust.dev/docs generally. No mention of informers, ONNX, or local/offline embedding inference for Ruby anywhere in Braintrust's documentation.
Upstream sources
Braintrust docs sources checked
Local repo files inspected
lib/braintrust/contrib.rb — registers only 4 integrations: OpenAI, RubyOpenAI, RubyLLM, Anthropic. No informers integration.lib/braintrust/contrib/ — no directory or files for informersAppraisals — no appraisal scenario for informersbraintrust.gemspec — no mention of informers- Grep for
informers, Informers, onnx across lib/braintrust/ returns zero matches
Summary
The
informersgem (RubyGems:informers, by Andrew Kane) is a Ruby library for running Hugging Face transformer models locally via ONNX Runtime — no API keys or network calls required. It provides a concrete execution surface for embeddings generation, text classification, and other transformer inference pipelines, and is not instrumented by this SDK.What is missing
No instrumentation exists for any
informersexecution surface. Key APIs that should be instrumented:Embeddings
Informers.pipeline("embedding", model_name).(text)— runs a local ONNX embedding model and returns vector output. This is the primary execution method analogous toembeddings.createin hosted embedding APIs.Other Inference Pipelines
Informers.pipeline("text-classification", ...),"feature-extraction","zero-shot-classification","reranking", etc. — each.callinvocation runs local model inference and is a natural span boundary (model name, input, output, latency).Weekly downloads
5.5K (as of 2026-07-23; https://clickgems.clickhouse.com/dashboard/informers) — ClickGems reports Last day / Last week / Last month / Total download counts sourced from Ruby Central's public RubyGems download dataset. RubyGems.org itself only exposes "Total downloads" and "downloads for this version," not a weekly figure, so ClickGems was used as the directly-reported weekly stat rather than deriving one.
Additional signals: v1.3.0 released April 15, 2026 (rubygems.org); Total downloads 2,515,045 (rubygems.org).
Braintrust docs status
not_found— Checked https://www.braintrust.dev/docs/integrations/sdk-integrations (lists onlyopenai,anthropic,ruby-llm,ruby-openaifor Ruby) and searched https://www.braintrust.dev/docs generally. No mention ofinformers, ONNX, or local/offline embedding inference for Ruby anywhere in Braintrust's documentation.Upstream sources
Braintrust docs sources checked
Local repo files inspected
lib/braintrust/contrib.rb— registers only 4 integrations: OpenAI, RubyOpenAI, RubyLLM, Anthropic. Noinformersintegration.lib/braintrust/contrib/— no directory or files forinformersAppraisals— no appraisal scenario forinformersbraintrust.gemspec— no mention ofinformersinformers,Informers,onnxacrosslib/braintrust/returns zero matches