AGENT SKILL

langchain-rate-limits

Implement LangChain rate limiting, retry strategies, and backoff. Use when handling API rate limits, controlling request throughput, or implementing concurrency-safe batch processing. Trigger: "langchain rate limit", "langchain throttling", "langchain backoff", "langchain retry", "API quota", "429 error".

License 许可
MIT
最近更新
2026-07-13
Stars
★ 1
Forks
0

01 前言 Front Matter

name
langchain-rate-limits
tags
[ "saas", "langchain", "api", "llm" ]
author
Jeremy Longshore <jeremy@intentsolutions.io>
license
MIT
version
1.0.0
description
Implement LangChain rate limiting, retry strategies, and backoff. Use when handling API rate limits, controlling request throughput, or implementing concurrency-safe batch processing. Trigger: "langchain rate limit", "langchain throttling", "langchain backoff", "langchain retry", "API quota", "429 error".
allowed-tools
Read, Write, Edit
compatible-with
claude-code, codex, openclaw

03 SKILL.md 正文

LangChain Rate Limits

Overview

Handle API rate limits gracefully with built-in retries, exponential backoff, concurrency control, provider fallbacks, and custom rate limiters.

Provider Rate Limits (2026)

Provider Model RPM TPM
OpenAI gpt-4o 10,000 800,000
OpenAI gpt-4o-mini 10,000 4,000,000
Anthropic claude-sonnet 4,000 400,000
Anthropic claude-haiku 4,000 400,000
Google gemini-1.5-pro 360 4,000,000

RPM = requests/minute, TPM = tokens/minute. Actual limits depend on your tier.

Strategy 1: Built-in Retry (Simplest)

import { ChatOpenAI } from "@langchain/openai";

// Built-in exponential backoff on 429/500/503
const model = new ChatOpenAI({
  model: "gpt-4o-mini",
  maxRetries: 5,      // retries with exponential backoff
  timeout: 30000,     // 30s timeout per request
});

// This automatically retries on rate limit errors
const response = await model.invoke("Hello");

Strategy 2: Concurrency-Controlled Batch

import { ChatOpenAI } from "@langchain/openai";
import { ChatPromptTemplate } from "@langchain/core/prompts";
import { StringOutputParser } from "@langchain/core/output_parsers";

const chain = ChatPromptTemplate.fromTemplate("Summarize: {text}")
  .pipe(new ChatOpenAI({ model: "gpt-4o-mini", maxRetries: 3 }))
  .pipe(new StringOutputParser());

const inputs = articles.map((text) => ({ text }));

// batch() with maxConcurrency prevents flooding the API
const results = await chain.batch(inputs, {
  maxConcurrency: 5,  // max 5 parallel requests
});

Strategy 3: Provider Fallback on Rate Limit

import { ChatOpenAI } from "@langchain/openai";
import { ChatAnthropic } from "@langchain/anthropic";

const primary = new ChatOpenAI({
  model: "gpt-4o-mini",
  maxRetries: 2,
  timeout: 10000,
});

const fallback = new ChatAnthropic({
  model: "claude-sonnet-4-20250514",
  maxRetries: 2,
});

// Automatically switches to Anthropic if OpenAI rate-limits
const resilientModel = primary.withFallbacks({
  fallbacks: [fallback],
});

const chain = prompt.pipe(resilientModel).pipe(new StringOutputParser());

Strategy 4: Custom Rate Limiter

class TokenBucketLimiter {
  private tokens: number;
  private lastRefill: number;

  constructor(
    private maxTokens: number,    // bucket size
    private refillRate: number,   // tokens per second
  ) {
    this.tokens = maxTokens;
    this.lastRefill = Date.now();
  }

  async acquire(): Promise<void> {
    this.refill();
    while (this.tokens < 1) {
      const waitMs = (1 / this.refillRate) * 1000;
      await new Promise((r) => setTimeout(r, waitMs));
      this.refill();
    }
    this.tokens -= 1;
  }

  private refill() {
    const now = Date.now();
    const elapsed = (now - this.lastRefill) / 1000;
    this.tokens = Math.min(this.maxTokens, this.tokens + elapsed * this.refillRate);
    this.lastRefill = now;
  }
}

// Usage: 100 requests per minute
const limiter = new TokenBucketLimiter(100, 100 / 60);

async function rateLimitedInvoke(chain: any, input: any) {
  await limiter.acquire();
  return chain.invoke(input);
}

Strategy 5: Async Batch with Semaphore

async function batchWithSemaphore<T>(
  chain: { invoke: (input: any) => Promise<T> },
  inputs: any[],
  maxConcurrent = 5,
): Promise<T[]> {
  let active = 0;
  const results: T[] = [];
  const queue = [...inputs.entries()];

  return new Promise((resolve, reject) => {
    function next() {
      while (active < maxConcurrent && queue.length > 0) {
        const [index, input] = queue.shift()!;
        active++;
        chain.invoke(input)
          .then((result) => {
            results[index] = result;
            active--;
            if (queue.length === 0 && active === 0) resolve(results);
            else next();
          })
          .catch(reject);
      }
    }
    next();
  });
}

// Process 100 items, 5 at a time
const results = await batchWithSemaphore(chain, inputs, 5);

Python Equivalent

from langchain_openai import ChatOpenAI
from langchain_anthropic import ChatAnthropic
from langchain_core.runnables import RunnableConfig

# Built-in retry
llm = ChatOpenAI(model="gpt-4o-mini", max_retries=5, request_timeout=30)

# Fallback
primary = ChatOpenAI(model="gpt-4o-mini", max_retries=2)
fallback = ChatAnthropic(model="claude-sonnet-4-20250514")
robust = primary.with_fallbacks([fallback])

# Batch with concurrency control
results = chain.batch(
    [{"text": t} for t in texts],
    config=RunnableConfig(max_concurrency=10),
)

Error Handling

Error Cause Fix
429 Too Many Requests Rate limit hit Increase maxRetries, reduce maxConcurrency
Timeout Response too slow Increase timeout, check network
QuotaExceeded Monthly limit hit Upgrade tier or switch provider
Batch partially fails Some items rate limited Use .batch() with returnExceptions: true

Resources

Next Steps

Proceed to langchain-security-basics for security best practices.

04 同仓库更多 Skills

jukeyman
jukeyman/jukeyman-skills

zustand-store-creator

Create zustand store creator operations. Auto-activating skill for Frontend Development. Triggers on: zustand store creator, zustand store creator Part of the Frontend Development skill category. Use when working with zustand store creator functionality. Trigger with phrases like "zustand store creator", "zustand creator", "zustand".

★ 1
jukeyman
jukeyman/jukeyman-skills

zero-trust-config-helper

Configure with zero trust config helper operations. Auto-activating skill for Security Advanced. Triggers on: zero trust config helper, zero trust config helper Part of the Security Advanced skill category. Use when configuring systems or services. Trigger with phrases like "zero trust config helper", "zero helper", "zero".

★ 1
jukeyman
jukeyman/jukeyman-skills

zapier-integration-helper

Assist with zapier integration helper operations. Auto-activating skill for Business Automation. Triggers on: zapier integration helper, zapier integration helper Part of the Business Automation skill category. Use when working with APIs or building integrations. Trigger with phrases like "zapier integration helper", "zapier helper", "zapier".

★ 1
jukeyman
jukeyman/jukeyman-skills

yaml-config-validator

Validate yaml config validator operations. Auto-activating skill for DevOps Basics. Triggers on: yaml config validator, yaml config validator Part of the DevOps Basics skill category. Use when configuring systems or services. Trigger with phrases like "yaml config validator", "yaml validator", "yaml".

★ 1
jukeyman
jukeyman/jukeyman-skills

xvary-stock-research

Thesis-driven equity analysis from public SEC EDGAR and market data; /analyze, /score, /compare workflows with bundled Python tools (Claude Code, Cursor, Codex).

★ 1
jukeyman
jukeyman/jukeyman-skills

xss-vulnerability-scanner

Scan xss vulnerability scanner operations. Auto-activating skill for Security Fundamentals. Triggers on: xss vulnerability scanner, xss vulnerability scanner Part of the Security Fundamentals skill category. Use when working with xss vulnerability scanner functionality. Trigger with phrases like "xss vulnerability scanner", "xss scanner", "xss".

★ 1