[bot] Cerebras Cloud TypeScript SDK (`@cerebras/cerebras_cloud_sdk`) not instrumented — no tracing for fast-inference chat completions or embeddings (116,524 weekly downloads)

Aberta
  • Algo errado?
#2,101 2 comentários 0 reações 0 responsáveis Ver no GitHub

Mantenedores costumam responder em até 1 dia

Ninguém assumiu esta issue ainda.

Avaliação

Dificuldade
4/5
Tempo estimado
3-5 dias
Facilidade para iniciantes
68/100
Tipo de issue
Funcionalidade
Clareza
Razoavelmente clara
Status de atividade
Ativa
Stack de tecnologia
node.js, typescript
Domínio
observability

Direção de pesquisa

Comece lendo as configurações existentes em js/src/auto-instrumentations/configs/, especialmente openai.ts, groq.ts e mistral.ts; depois, inspecione js/src/wrappers/ e js/src/instrumentation/plugins/ em busca dos padrões correspondentes. Adicione cobertura para Cerebras chat completions, streaming, completions e embeddings, com uma configuração de auto-instrumentation e um cenário e2e em e2e/scenarios/; considera-se concluído quando chamadas diretas a @cerebras/cerebras_cloud_sdk produzirem spans do Braintrust.

Escrita pelo modelo de indexação a partir do texto da issue.

Descrição

Summary

Cerebras publishes an official TypeScript SDK for their ultra-fast LLM inference API: @cerebras/cerebras_cloud_sdk on npm (v1.64.1, actively maintained). It provides execution APIs for chat completions, completions, embeddings, and streaming responses. The SDK has its own HTTP client based on node-fetch — it does not depend on the openai npm package. This repository has zero instrumentation for any Cerebras SDK surface — no wrapper, no channels, no plugin, no auto-instrumentation config. Users who call @cerebras/cerebras_cloud_sdk directly get no Braintrust spans.

What instrumentation is missing

The @cerebras/cerebras_cloud_sdk npm package exposes these execution surfaces, none of which are instrumented:

SDK Method Description
client.chat.completions.create({ model, messages }) Chat completions with tool use, structured outputs, and reasoning
client.chat.completions.create({ ..., stream: true }) Streaming chat completions
client.completions.create({ model, prompt }) Raw text completions
client.embeddings.create(...) Text embeddings

Cerebras is notable for extremely low-latency LLM inference (their wafer-scale hardware delivers tokens significantly faster than GPU-based providers), making observability especially valuable for latency-sensitive production workloads.

No coverage in any instrumentation layer:

  • No wrapper function (e.g. wrapCerebras())
  • No diagnostics channels for Cerebras methods
  • No plugin handler in js/src/instrumentation/plugins/
  • No auto-instrumentation config in js/src/auto-instrumentations/configs/
  • No e2e test scenarios
  • Grep for cerebras (case-insensitive) across js/src/ returns zero matches

Indirect coverage is insufficient:

Braintrust's Cerebras integration page documents a gateway/OpenAI SDK approach only — using the openai npm client pointed at the Braintrust gateway to route to Cerebras models. Users who call @cerebras/cerebras_cloud_sdk directly get no spans at all. The Cerebras SDK has its own separate HTTP layer (runtime deps: node-fetch, abort-controller, agentkeepalive, form-data-encoder, formdata-node) and is not patched by the existing openai auto-instrumentation.

Braintrust docs status

not_found for direct SDK instrumentation. The Braintrust Cerebras integration page documents a gateway/OpenAI SDK approach only and does not reference @cerebras/cerebras_cloud_sdk.

Upstream references

Local files inspected

  • js/src/auto-instrumentations/configs/ — full config list: anthropic.ts, openai.ts, ai-sdk.ts, groq.ts, mistral.ts, cohere.ts, huggingface.ts, google-genai.ts, google-adk.ts, genkit.ts, openai-agents.ts, openrouter.ts, openrouter-agent.ts, claude-agent-sdk.ts, cursor-sdk.ts, github-copilot.ts, openai-codex.ts, flue.ts — no Cerebras config present
  • js/src/wrappers/ — no cerebras.ts wrapper file
  • js/src/instrumentation/plugins/ — no Cerebras channels or plugin
  • e2e/scenarios/ — no Cerebras test scenarios
  • Full repo grep for cerebras (case-insensitive) in js/src/ — zero matches
  • Cerebras SDK package.json (GitHub) — confirmed runtime deps are node-fetch/agentkeepalive/form-data-encoder; no openai dependency

Weekly downloads: 116,524 (as of 2026-10-14; window 2026-10-06 to 2026-10-12; source https://api.npmjs.org/downloads/point/last-week/@cerebras/cerebras_cloud_sdk)

Linguagem predominante
TypeScript
Estrelas
29
Forks
22
Merge médio
1d 25min
PRs com merge (30d)
80

Preparar o ambiente

  • Inclui um Dockerfile ou arquivo Docker Compose
  • Sem modelo de pull request
  • Sem guia de contribuição

Primeiros passos

  1. Leia a issue inteira e depois o guia de contribuição do projeto.
  2. Comente na issue dizendo que vai assumir — evita que duas pessoas façam o mesmo trabalho.
  3. Faça um fork do repositório e trabalhe em uma branch.
  4. Abra um pull request que referencie o número da issue.

Mais de braintrustdata/braintrust-sdk-javascript

Todas as issues de braintrustdata/braintrust-sdk-javascript

Issues semelhantes

Mais issues de TypeScript

Receba novas issues na sua caixa de entrada

Um resumo curto de issues do GitHub para quem está começando.