Nyumbani / Miongozo / API ya LLM ya bure kwa LangChain

API ya LLM ya bure kwa LangChain

LangChain inaunganishwa na API zinazooana na OpenAI kupitia ChatOpenAI na OpenAIEmbeddings. Unabadilisha base URL na ufunguo. Mwongozo huu unaelekeza madarasa yote mawili kwenye FreeLLMAPI, ambayo inakusanya ngazi 45 za bure za watoa huduma nyuma ya ufunguo mmoja. Msimbo ulio hapa chini unafanya kazi kwa Python na JavaScript na unashughulikia vikomo vya kiwango.

Msimbo, majina ya modeli na lebo za katalogi ya moja kwa moja zinaonyeshwa kwa Kiingereza.

Sanidi lango na vifurushi

Sakinisha FreeLLMAPI kisha ongeza funguo za bure kutoka watoa huduma wawili au zaidi kwenye ukurasa wa Keys wa dashibodi. Nakili ufunguo wa pamoja. Msimbo wako unatumia ufunguo huu pekee. Funguo za watoa huduma zinabaki ndani ya FreeLLMAPI.

Sakinisha FreeLLMAPI na vifurushi vya LangChain

curl -fsSL https://freellmapi.co/install.sh | bash

# Python
pip install -U langchain langchain-openai

# JavaScript
npm install @langchain/openai @langchain/core zod

Vigezo vya mazingira

export FREELLMAPI_BASE_URL=http://localhost:3001/v1
export FREELLMAPI_API_KEY=<unified-key>

Tumia model="auto" ili kipanga njia kichague modeli. Unaweza pia kutumia mkakati wa uelekezaji, kwa mfano auto:fast au auto:smart. Wasifu wenye jina kama auto:coding na model ID isiyobadilika kutoka katalogi pia zinafanya kazi.

Ili kuona modeli ambazo funguo zako zinaweza kuita sasa, omba GET /v1/models?execution_status=ready. Orodha kamili pia ina modeli zisizo na ufunguo. Modeli yenye hali exhausted ina funguo, lakini zote zimefikia kikomo kwa sasa. Hali hiyo inaisha muda wa kupumzika unapomalizika.

Orodhesha modeli zinazoweza kujibu sasa

curl "http://localhost:3001/v1/models?execution_status=ready" \
  -H "Authorization: Bearer <unified-key>"

Gumzo kwa ChatOpenAI

Kwenye Python, weka base_url na api_key. Thamani ya max_retries inakwenda kwa mteja wa OpenAI, ambaye anajaribu tena makosa 429 na 5xx kwa kusubiri kwa ongezeko la kielelezo.

Python: gumzo na utiririshaji

import os
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="auto",
    base_url=os.environ["FREELLMAPI_BASE_URL"],
    api_key=os.environ["FREELLMAPI_API_KEY"],
    temperature=0.2,
    max_retries=3,
    timeout=60,
    include_response_headers=True,
)

msg = llm.invoke("Explain retrieval-augmented generation in two sentences.")
print(msg.content)
print("Served by:", msg.response_metadata["headers"].get("x-routed-via"))

for chunk in llm.stream("Write a haiku about SQLite."):
    print(chunk.content, end="", flush=True)

Kwenye JavaScript, weka base URL kwenye configuration.baseURL. Kitu cha configuration kinakwenda kwa mteja rasmi wa OpenAI SDK.

JavaScript: gumzo na utiririshaji

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

const llm = new ChatOpenAI({
  model: "auto",
  apiKey: process.env.FREELLMAPI_API_KEY,
  temperature: 0.2,
  maxRetries: 3,
  configuration: { baseURL: process.env.FREELLMAPI_BASE_URL },
});

const msg = await llm.invoke("Explain retrieval-augmented generation in two sentences.");
console.log(msg.content);

for await (const chunk of await llm.stream("Write a haiku about SQLite.")) {
  process.stdout.write(String(chunk.content));
}

Kichwa cha x-routed-via kinaonyesha mtoa huduma na modeli waliojibu. Ikiwa kipanga njia kimebadilisha mtoa huduma, x-fallback-trail inaonyesha kila jaribio lililoshindwa na sababu yake.

Zana na mawakala

FreeLLMAPI inapeleka tools za mtindo wa OpenAI kwa kila mtoa huduma na kurudisha tool_calls katika muundo wa OpenAI. Kwa Gemini, inabadilisha ombi kuwa muundo wa Google na kurudi tena. Tumia bind_tools kwa wito mmoja, au create_agent kwa mzunguko unaoendesha zana.

Python: bind_tools na create_agent

from langchain.agents import create_agent
from langchain_core.tools import tool

@tool
def get_weather(city: str) -> str:
    """Return the current weather for a city."""
    return f"It is 18 C and cloudy in {city}."

# One call: the model returns the tool call, your code runs it
msg = llm.bind_tools([get_weather]).invoke("What is the weather in Dublin?")
print(msg.tool_calls)

# Agent loop: LangChain runs the tool and sends the result back
agent = create_agent(
    model=llm,
    tools=[get_weather],
    system_prompt="You answer weather questions. Use the tool.",
)
result = agent.invoke({"messages": [{"role": "user", "content": "Weather in Dublin?"}]})
print(result["messages"][-1].content)

JavaScript: bindTools

import { tool } from "@langchain/core/tools";
import { z } from "zod";

const getWeather = tool(
  async ({ city }) => `It is 18 C and cloudy in ${city}.`,
  {
    name: "get_weather",
    description: "Return the current weather for a city.",
    schema: z.object({ city: z.string() }),
  },
);

const reply = await llm.bindTools([getWeather]).invoke("What is the weather in Dublin?");
console.log(reply.tool_calls);

Baadhi ya modeli za bure pekee zinaunga mkono wito wa zana. Kwa auto, weka modeli zinazoweza kuita zana kwenye mnyororo wako wa akiba, au tengeneza wasifu wa mawakala. Modeli hizi za bure zinaunga mkono zana kwa sasa:

Modeli za bure zenye wito wa zana na muktadha wa 128K au zaidi
ModeliBure kwenyeMuktadhaVikomo vya bure (ofa bora)ZanaUoni
GLM 5.3UnoRouter1M1 maombi/dakika
1 req/min per model
NdiyoHapana
GLM 5.3 FlashNVIDIA NIM, UnoRouter1M40 maombi/dakika
40 RPM
NdiyoNdiyo
Gemini 3.8 FlashGoogle AI Studio1MRecurring provider quotaNdiyoNdiyo
Qwen3.8 Flash NextAMD Radeon Cloud262KRecurring daily account allowanceNdiyoHapana
Qwen3.8 FlashElectronHub1M$0.25/week shared recurring creditNdiyoNdiyo
Gemini 3.7 FlashGoogle AI Studio1M10 maombi/dakika
20 maombi/siku
250,000 tokeni/dakika
~3M
NdiyoNdiyo
GPT-5.6 LunaElectronHub1M$0.25/week shared recurring creditNdiyoNdiyo
Dots3 Note PreviewKilo Gateway, OpenRouter512K20 maombi/dakika
50 maombi/siku
50 req/day account-wide (1,000 after a $10 purchase)
NdiyoNdiyo
DeepSeek V4 FlashAMD Radeon Cloud, BazaarLink, ElectronHub1M10 maombi/dakika
50 maombi/siku
Shared rate limit
NdiyoHapana
Qwen3.8 27BElectronHub, Groq, OVH AI Endpoints262K30 maombi/dakika
250 maombi/siku
70,000 tokeni/dakika
~6M
NdiyoNdiyo

Inaonyesha 10 kati ya 109 modeli za bure zinazolingana. Tazama modeli zote.

Data ya moja kwa moja kutoka katalogi ya FreeLLMAPI (toleo 2026.11.19, imekaguliwa 2026-11-19). Vikomo vya bure hubadilika mara kwa mara. Katalogi hukaguliwa kila siku.

Embeddings kwa OpenAIEmbeddings

Kwa chaguo-msingi, OpenAIEmbeddings kwenye Python inagawanya maandishi kwa kitenganishi cha tokeni cha OpenAI na kutuma tokeni. Watoa huduma wengi wengine wanahitaji maandishi ghafi. Weka check_embedding_ctx_length=False ili kutuma maandishi.

Python: embeddings

from langchain_openai import OpenAIEmbeddings

emb = OpenAIEmbeddings(
    model="bge-m3",  # one embedding family, 1024 dimensions
    base_url=os.environ["FREELLMAPI_BASE_URL"],
    api_key=os.environ["FREELLMAPI_API_KEY"],
    check_embedding_ctx_length=False,
)
vectors = emb.embed_documents(["FreeLLMAPI pools free tiers.", "LangChain builds LLM apps."])
print(len(vectors), len(vectors[0]))

JavaScript: embeddings

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

const emb = new OpenAIEmbeddings({
  model: "bge-m3",
  apiKey: process.env.FREELLMAPI_API_KEY,
  configuration: { baseURL: process.env.FREELLMAPI_BASE_URL },
});
const vector = await emb.embedQuery("What does FreeLLMAPI do?");
console.log(vector.length);

Mbadala wa embedding kwenye FreeLLMAPI haubadilishi modeli kamwe. Unahamia tu kati ya watoa huduma wanaotoa familia ile ile. Hii inaweka vekta zote kwenye hifadhi moja ya vekta zikiwa sambamba. Chagua familia moja kisha ibaki nayo. Modeli za bure za embedding za leo ni:

ModeliMtoa hudumaVipimoIngizo la juu zaidiKikomo cha bure
Gemini Embedding
gemini-embedding-001
Google AI Studio30722048100 rpm · 1K req/day
Nemotron Embed VL 1B
nvidia/llama-nemotron-embed-vl-1b-v2
NVIDIA NIM20488192~40 rpm
Nemotron 3 Embed 1B (OR free)
nvidia/nemotron-3-embed-1b:free
OpenRouter20483276820 rpm · shared free daily cap
SEA-LION E5 Embedding 600M
aisingapore/SEA-LION-E5-Embedding-600M
SEA-LION10248192free · 10 rpm
SEA-LION ModernBERT Embedding 300M
aisingapore/SEA-LION-ModernBERT-Embedding-300M
SEA-LION10248192free · 10 rpm
SEA-LION ModernBERT Embedding 600M
aisingapore/SEA-LION-ModernBERT-Embedding-600M
SEA-LION10248192free · 10 rpm
BGE-M3
@cf/baai/bge-m3
Cloudflare Workers AI1024819210K neurons/day (shared)
EmbeddingGemma 300M
@cf/google/embeddinggemma-300m
Cloudflare Workers AI768204810K neurons/day (shared)
Qwen3 Embedding 0.6B
@cf/qwen/qwen3-embedding-0.6b
Cloudflare Workers AI1024819210K neurons/day (shared)
Cohere Embed v4
embed-v4.0
Cohere15361280001K calls/mo (shared w/ chat)
Gemini Embedding 2
gemini-embedding-2
Google AI Studio30728192free-tier rate limited
Nemotron 3 Embed 1B
nvidia/nemotron-3-embed-1b
NVIDIA NIM20484096~40 rpm
PLaMo Embedding 1B (CF)
@cf/pfnet/plamo-embedding-1b
Cloudflare Workers AI2048409610K neurons/day (shared)
BGE Small EN v1.5 (CF)
@cf/baai/bge-small-en-v1.5
Cloudflare Workers AI38451210K neurons/day (shared)
BGE Base EN v1.5 (CF)
@cf/baai/bge-base-en-v1.5
Cloudflare Workers AI76851210K neurons/day (shared)
BGE Large EN v1.5 (CF)
@cf/baai/bge-large-en-v1.5
Cloudflare Workers AI102451210K neurons/day (shared)
Mistral Embed
mistral-embed
Mistral10248192free mode · monthly cap (shared)
Codestral Embed
codestral-embed
Mistral15368192free mode · monthly cap (shared)
Llama Text Embed v2 (Pinecone)
llama-text-embed-v2
Pinecone102420485M tokens/mo per model
Multilingual E5 Large (Pinecone)
multilingual-e5-large
Pinecone10245075M tokens/mo per model
Nemotron 3 Embed 1B (Speka)
nvidia/nemotron-3-embed-1b
Speka20488192$1/month shared across Speka models
OpenAI: Text Embedding 3 Small (ElectronHub)
text-embedding-3-small
ElectronHub15368192$0.25/week shared free credits
OpenAI: Text Embedding 3 Large (ElectronHub)
text-embedding-3-large
ElectronHub30728192$0.25/week shared free credits
Nemotron 3 Embed 8B (Inferbase)
nvidia-nemotron-3-embed-8b
Inferbase409681925,000 requests/month; 100 requests and $0.50 usage/day shared
Qwen 3 Embedding 0.6B (Inferbase)
qwen3-embedding-0-6b
Inferbase102481925,000 requests/month; 100 requests and $0.50 usage/day shared
Qwen 3 Embedding 4B (Inferbase)
qwen3-embedding-4b
Inferbase256081925,000 requests/month; 100 requests and $0.50 usage/day shared
Qwen 3 Embedding 8B (Inferbase)
qwen3-embedding-8b
Inferbase409681925,000 requests/month; 100 requests and $0.50 usage/day shared
LFM2.5 Embedding 350M (LLMTR)
liquid/lfm-2.5-embedding-350m-free
LLMTR1024512Ongoing zero-price route; provider terms apply
BGE-M3 (SEA-LION)
BAAI/bge-m3
SEA-LION10248192free · 10 rpm
Cohere Embed English v3
embed-english-v3.0
Cohere10245121K calls/mo (shared w/ chat)
Qwen3 Embedding 8B (OVH)
Qwen3-Embedding-8B
OVH AI Endpoints409632768free · anonymous 2 req/min per IP
BGE-M3 (OVH)
bge-m3
OVH AI Endpoints10248192free · anonymous 2 req/min per IP
BGE Multilingual Gemma2 (OVH)
bge-multilingual-gemma2
OVH AI Endpoints35848192free · anonymous 2 req/min per IP
Mistral: Codestral Embed (ElectronHub)
codestral-embed
ElectronHub15368192$0.25/week shared free credits
Mistral: Mistral Embed (ElectronHub)
mistral-embed
ElectronHub10248192$0.25/week shared free credits
Google: Gemini Embedding 2 (ElectronHub)
gemini-embedding-2
ElectronHub30728192$0.25/week shared free credits
Google: Gemini Embedding 001 (ElectronHub)
gemini-embedding-001
ElectronHub30722048$0.25/week shared free credits
OpenAI: Text Embedding Ada 002 (ElectronHub)
text-embedding-ada-002
ElectronHub15368192$0.25/week shared free credits
Cohere Embed English Light v3
embed-english-light-v3.0
Cohere3845121K calls/mo (shared w/ chat)
Cohere Embed Multilingual v3
embed-multilingual-v3.0
Cohere10245121K calls/mo (shared w/ chat)
Cohere Embed Multilingual Light v3
embed-multilingual-light-v3.0
Cohere3845121K calls/mo (shared w/ chat)

Data ya moja kwa moja kutoka katalogi ya FreeLLMAPI (toleo 2026.11.19, imekaguliwa 2026-11-19). Vikomo vya bure hubadilika mara kwa mara. Katalogi hukaguliwa kila siku.

Matokeo yaliyopangwa, vikao na uchaguzi wa modeli

with_structured_output inarudisha kitu cha Pydantic badala ya maandishi. Kwa chaguo-msingi, LangChain inatumia wito wa kazi kwa hili. Mbinu hiyo inafanya kazi kwenye kila modeli ya bure inayounga mkono zana.

Python: matokeo yaliyopangwa

from pydantic import BaseModel, Field

class Ticket(BaseModel):
    category: str = Field(description="billing, bug or other")
    urgent: bool

classifier = llm.with_structured_output(Ticket)
ticket = classifier.invoke("The app crashes when I pay. Please fix it today.")
print(ticket.category, ticket.urgent)

Usitumie method="json_schema" na auto. Si kila mtoa huduma wa bure anaunga mkono matokeo madhubuti ya JSON schema, na kipanga njia kinaweza kuchagua mtoa huduma tofauti kwa kila ombi.

FreeLLMAPI inaweka mazungumzo kwenye modeli moja kwa dakika 30. Inatambua mazungumzo kutoka kwenye kichwa cha X-Session-Id. Ikiwa hakuna kichwa, inatumia hash ya ujumbe wa kwanza wa mtumiaji. Kwa programu ya gumzo yenye watumiaji wengi, tuma kitambulisho kimoja cha kikao kwa kila mtumiaji:

Python: kikao kimoja kwa kila mtumiaji

def llm_for(user_id: str) -> ChatOpenAI:
    return ChatOpenAI(
        model="auto",
        base_url=os.environ["FREELLMAPI_BASE_URL"],
        api_key=os.environ["FREELLMAPI_API_KEY"],
        default_headers={"X-Session-Id": f"user-{user_id}"},
        max_retries=3,
    )

Chagua mfuatano wa modeli kwa kila sehemu ya programu yako:

Kazi kwenye programu yakoMfuatano wa modeliSababu
Majibu ya gumzo kwa watumiajiautoInafuata mnyororo ulioweka kwenye dashibodi
Wakala mwenye zanaauto:<profile>Wasifu wenye modeli za wito wa zana pekee
Uainishaji, uelekezaji, muhtasari mfupiauto:fastInapanga modeli kwa kasi iliyopimwa
Hatua ngumu za kufikiriauto:smartInapanga modeli kwa uwezo wa akili
Majaribio yanayopaswa kurudiwa kwa matokeo yaleyaleModel ID isiyobadilikaModeli ile ile kila wakati, lakini inasimama kwenye kikomo cha modeli hiyo

Ngazi za bure hazina SLA. Muda wa kujibu hubadilika wakati wa siku, na modeli zenye nguvu zaidi hufikia vikomo vyao vya kila siku kwanza. Jaribu minyororo yako kwa modeli ambazo mnyororo wako utazitumia kweli.

Vikomo vya kiwango na majaribio upya

FreeLLMAPI inajaribu mtoa huduma anayefuata mtoa huduma mmoja anaporudisha 429. Msimbo wako unaona 429 pale tu funguo zote kwenye mnyororo zinapofikia kikomo chake. Kisha jibu lina kichwa cha Retry-After. Vikomo vingi vya kila siku huwekwa upya saa sita usiku UTC.

Kwa kazi za makundi, ongeza tabaka la pili la majaribio upya kwa with_retry. Inajaribu upya runnable nzima kwa kusubiri kwa ongezeko la kielelezo na jitter.

Python: jaribu upya mnyororo kwenye vikomo vya kiwango

import openai
from langchain_core.prompts import ChatPromptTemplate

prompt = ChatPromptTemplate.from_messages([
    ("system", "Summarize the text in one sentence."),
    ("user", "{text}"),
])

chain = (prompt | llm).with_retry(
    retry_if_exception_type=(openai.RateLimitError, openai.APITimeoutError),
    wait_exponential_jitter=True,
    stop_after_attempt=5,
)

# batch() runs requests in parallel; max_concurrency keeps the rate low
results = chain.batch(
    [{"text": t} for t in ["First document.", "Second document."]],
    config={"max_concurrency": 4},
)
TatizoChanzoSuluhisho
openai.AuthenticationErrorUfunguo wa pamoja si sahihiNakili tena ufunguo kutoka ukurasa wa Keys
openai.NotFoundError kwa kila witoBase URL haina /v1Tumia http://localhost:3001/v1
openai.RateLimitErrorFunguo zote kwenye mnyororo zimefikia kikomo chakeOngeza watoa huduma, punguza max_concurrency au subiri Retry-After
Hitilafu ya embedding kuhusu tokeni au aina ya ingizoLangChain inatuma tokeni, si maandishiWeka check_embedding_ctx_length=False
tool_calls ni tupuModeli haiungi mkono zanaTumia modeli kutoka jedwali la zana au wasifu wa zana
Maandishi ya kufikiri hayapoChatOpenAI inahifadhi sehemu za kawaida za OpenAI pekeeSoma jibu kwenye content; sehemu za ziada za mtoa huduma hazihifadhiwi
Tumia kila moja ya hizi kwa ufunguo mmoja. FreeLLMAPI ni kipanga njia huria, unachoendesha mwenyewe. Kinaweka tabaka hizi zote za bure nyuma ya kituo kimoja cha OpenAI kinachooana, na kinahamia kwenye nyingine pale kikomo kinapofikiwa. Vinjari katalogi au anza sasa.

Maswali yanayoulizwa mara kwa mara

Je, ninaweza kutumia API ya LLM ya bure na LangChain?

Ndiyo. API yoyote inayooana na OpenAI inafanya kazi na ChatOpenAI. Weka base_url kuwa http://localhost:3001/v1 na api_key kuwa ufunguo wako wa pamoja wa FreeLLMAPI. Kipanga njia kinatumia ngazi za bure za watoa huduma.

Ninawekaje base URL maalum kwenye LangChain ChatOpenAI?

Kwenye Python, pitisha base_url kwa ChatOpenAI. Kwenye JavaScript, pitisha configuration: { baseURL } kwa constructor. Kigezo cha mazingira cha OPENAI_BASE_URL pia kinafanya kazi kwenye Python.

Je, mawakala wa LangChain wanafanya kazi na modeli za bure?

Ndiyo, ikiwa modeli inaunga mkono wito wa zana. Tumia create_agent na modeli ya ChatOpenAI, kisha weka modeli zinazoweza kuita zana kwenye mnyororo wako wa FreeLLMAPI.

Kwa nini OpenAIEmbeddings inashindwa na mtoa huduma asiye wa OpenAI?

Kwa chaguo-msingi inatuma vitambulisho vya tokeni. Watoa huduma wengi wengine wanahitaji maandishi ghafi. Weka check_embedding_ctx_length=False kwenye Python.

Ninashughulikiaje hitilafu za 429 kwenye LangChain?

Weka max_retries kwenye ChatOpenAI na ongeza with_retry kwenye mnyororo kwa kazi za makundi. FreeLLMAPI tayari inahamia kwa mtoa huduma anayefuata kabla msimbo wako haujaona 429.

Miongozo inayohusiana

Vyanzo

Miongozo yote · Tazama modeli zote za bure