API ya LLM ya bure kwa Open WebUI
Open WebUI ni kiolesura cha gumzo unachoendesha mwenyewe. Kinaunganishwa na seva yoyote inayotumia API ya OpenAI Chat Completions. Mwongozo huu unaunganisha Open WebUI na FreeLLMAPI, ambayo inakusanya ngazi 45 za bure za watoa huduma nyuma ya ufunguo mmoja. Unaongeza muunganisho mmoja. Kipanga njia kinatuma kila gumzo kwenye modeli ya bure, na kinahamia kwenye mtoa huduma anayefuata kikomo kinapozuia ombi.
Msimbo, majina ya modeli na lebo za katalogi ya moja kwa moja zinaonyeshwa kwa Kiingereza.
Muunganisho unavyofanya kazi
Open WebUI inatuma maombi ya gumzo kwenye URL ya kila muunganisho wa OpenAI. FreeLLMAPI inasikiliza kwenye http://localhost:3001/v1 na inakubali maombi yale yale. Inachagua modeli ya bure, inatuma ombi na kurudisha jibu katika muundo wa OpenAI.
- Muunganisho mmoja, watoa huduma wengi: unaongeza URL moja na ufunguo mmoja wa pamoja. Funguo zako za watoa huduma zinabaki ndani ya FreeLLMAPI kwenye kompyuta yako.
- Mbadala: mtoa huduma anaporudisha hitilafu 429, kipanga njia kinajaribu modeli inayofuata kwenye mnyororo wako wa akiba.
- Vikomo: ngazi za bure zina vikomo vya maombi na tokeni. Funguo nyingi za watoa huduma zinatoa maombi zaidi kwa siku. Kuhamia kwa mbadala kunasambaza mzigo, lakini hakuondoi vikomo.
Sakinisha FreeLLMAPI kwanza. Inaendeshwa kwenye Docker. Fungua dashibodi kwenye http://localhost:3001, ongeza angalau funguo mbili za watoa huduma za bure kwenye ukurasa wa Keys, kisha nakili ufunguo wako wa pamoja.
Sakinisha FreeLLMAPI (macOS au Linux)
curl -fsSL https://freellmapi.co/install.sh | bashHatua ya 1: Pata URL sahihi
URL ambayo Open WebUI lazima itumie inategemea mahali kila programu inapoendeshwa. Open WebUI mara nyingi inaendeshwa kwenye Docker. Ndani ya kontena, localhost ni kontena lenyewe, si kompyuta yako.
| Open WebUI inaendeshwa | FreeLLMAPI inaendeshwa | URL ya kutumia |
|---|---|---|
| Kwenye mashine mwenyeji (pip au uv) | Kwenye mashine mwenyeji (Docker au programu ya eneo-kazi) | http://localhost:3001/v1 |
| Kwenye Docker Desktop (macOS, Windows) | Kwenye mashine mwenyeji | http://host.docker.internal:3001/v1 |
| Kwenye Docker kwenye Linux | Kwenye Docker (kwa hati ya usakinishaji) | http://freellmapi:3001/v1 kwenye mtandao wa pamoja wa Docker |
| Kwenye kompyuta nyingine | Kwenye seva ya mtandao wako wa ndani (LAN) | http://<server-ip>:3001/v1, pamoja na HOST_BIND=0.0.0.0 kwenye mtandao unaoaminika pekee |
Kwa chaguo-msingi, kontena la FreeLLMAPI linaweka mlango wake kwenye 127.0.0.1 pekee. Kwenye Linux, kontena la pili halifikii mlango huo kupitia mashine mwenyeji kila wakati. Unganisha kontena zote mbili kwenye mtandao mmoja wa Docker. Kisha tumia jina la huduma freellmapi kama jina la mwenyeji.
Unganisha Open WebUI kwenye mtandao wa FreeLLMAPI (Linux)
# Find the network of the FreeLLMAPI stack (usually freellmapi_default)
docker network ls
# Attach the Open WebUI container to it
docker network connect freellmapi_default open-webui
# Test the URL from inside the Open WebUI container
docker exec open-webui curl -s http://freellmapi:3001/v1/models \
-H "Authorization: Bearer <unified-key>" | head -c 300Hatua ya 2: Ongeza muunganisho
Unaweza kuongeza muunganisho kwenye kiolesura au kwa vigezo vya mazingira. Kiolesura ndiyo njia rahisi zaidi.
- Fungua Admin Panel kisha nenda kwenye Settings > Connections.
- Tafuta orodha ya Manage OpenAI API Connections. Bofya Add Connection.
- Kwenye URL, andika URL kutoka Hatua ya 1, kwa mfano
http://host.docker.internal:3001/v1. - Kwenye API Key, bandika ufunguo wako wa pamoja wa FreeLLMAPI.
- Bofya Verify Connection. Kisha bofya Save.
Ili kuweka muunganisho kontena linapoanza, tumia OPENAI_API_BASE_URL na OPENAI_API_KEY. Amri iliyo hapa chini ni amri ya kawaida ya docker run ya Open WebUI, ikiwa na vigezo hivi viwili vimeongezwa.
docker run yenye muunganisho wa FreeLLMAPI
docker run -d -p 3000:8080 \
--add-host=host.docker.internal:host-gateway \
-e OPENAI_API_BASE_URL=http://host.docker.internal:3001/v1 \
-e OPENAI_API_KEY=<unified-key> \
-v open-webui:/app/backend/data \
--name open-webui --restart always \
ghcr.io/open-webui/open-webui:mainIli kuhifadhi muunganisho wa OpenAI uliopo, tumia vigezo vya wingi. OPENAI_API_BASE_URLS na OPENAI_API_KEYS zinapokea orodha zilizotenganishwa kwa semikoloni. Ufunguo wa kwanza unakwenda na URL ya kwanza.
Miunganisho miwili kwenye kontena moja
-e OPENAI_API_BASE_URLS="https://api.openai.com/v1;http://host.docker.internal:3001/v1" \
-e OPENAI_API_KEYS="<openai-key>;<unified-key>"ENABLE_PERSISTENT_CONFIG=False ili mazingira yashinde.Hatua ya 3: Weka orodha fupi ya modeli
FreeLLMAPI inarudisha katalogi yake yote kwenye /v1/models. Katalogi pia ina modeli ambazo huwezi kuzitumia bado, kwa sababu hujaongeza ufunguo wake. Open WebUI inaonyesha kila modeli ambayo muunganisho unarudisha. Orodha ndefu ya modeli zinazoshindwa haina faida.
Fungua muunganisho tena kisha tumia sehemu ya Model IDs. Unapoongeza IDs, Open WebUI inaonyesha IDs hizi pekee na inaacha kusoma orodha yote. Ongeza ingizo za kipanga njia kwanza:
| Model ID | Kipanga njia kinafanya nini |
|---|---|
auto | Inafuata mnyororo wako wa akiba unaotumika kutoka dashibodi |
auto:fast | Inapanga modeli zilizowezeshwa kwa kasi iliyopimwa |
auto:smart | Inapanga modeli zilizowezeshwa kwa uwezo wa akili |
auto:reliable | Inapanga modeli zilizowezeshwa kwa kiwango cha mafanikio cha hivi karibuni |
auto:<profile> | Inatumia mnyororo wa akiba wenye jina, kwa mfano mnyororo wa uoni |
Unaweza pia kuongeza model ID isiyobadilika, kwa mfano modeli kutoka jedwali hapa chini. Modeli isiyobadilika inatoa jibu la mtindo uleule kila mara. Pia inasimama modeli hiyo inapomaliza kikomo chake cha bure, isipokuwa modeli hiyo hiyo ni ya bure kwa watoa huduma wengine.
Ili kuokoa kiwango, weka modeli ya kazi kuwa ingizo la haraka. Open WebUI inatumia modeli ya kazi kwa vichwa vya habari na maswali ya utafutaji. Kigezo ni TASK_MODEL_EXTERNAL, kwa mfano TASK_MODEL_EXTERNAL=auto:fast.
| Modeli | Bure kwenye | Muktadha | Vikomo vya bure (ofa bora) | Zana | Uoni |
|---|---|---|---|---|---|
| GLM 5.3 | APINex, UnoRouter | 1M | 5 maombi/dakika 10M weighted tokens/day shared across APINex free routes; 4x token multiplier; 5 requests/minute/account | Ndiyo | Hapana |
| GLM 5.3 Flash | APINex, BlazeAPI, NVIDIA NIM +1 zaidi | 1M | 40 maombi/dakika 40 RPM | Ndiyo | Ndiyo |
| Gemini 3.8 Flash | APINex, Google AI Studio | 1M | 5 maombi/dakika 10M weighted tokens/day shared across APINex free routes; 3x token multiplier; 5 requests/minute/account | Ndiyo | Ndiyo |
| Qwen3.8 Flash Next | AMD Radeon Cloud, BlazeAPI, ElectronHub | 262K | Recurring daily account allowance | Ndiyo | Hapana |
| Qwen3.8 Flash | ElectronHub | 1M | $0.25/week shared recurring credit | Ndiyo | Ndiyo |
| Deepseek V4.1 Flash | APINex | 1M | 5 maombi/dakika 10M weighted tokens/day shared across APINex free routes; 3x token multiplier; 5 requests/minute/account | Hapana | Hapana |
| Gemini 3.7 Flash | Google AI Studio | 1M | 10 maombi/dakika 20 maombi/siku 250,000 tokeni/dakika ~3M | Ndiyo | Ndiyo |
| GPT-5.6 Luna | ElectronHub | 1M | $0.25/week shared recurring credit | Ndiyo | Ndiyo |
| Dots3 Note Preview | Kilo Gateway, OpenRouter | 512K | 20 maombi/dakika 50 maombi/siku 50 req/day account-wide (1,000 after a $10 purchase) | Ndiyo | Ndiyo |
| DeepSeek V4 Flash | AMD Radeon Cloud, BazaarLink, ElectronHub | 1M | 10 maombi/dakika 50 maombi/siku Shared rate limit | Ndiyo | Hapana |
Inaonyesha 10 kati ya 184 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.
Ukituma picha kwenye Open WebUI, kipanga njia kinatumia modeli za uoni kwa ombi hilo pekee. Ikiwa hakuna modeli ya uoni kwenye mnyororo wako, ombi linarudisha hitilafu 422. Modeli hizi za bure zinakubali picha:
| Modeli | Bure kwenye | Muktadha | Vikomo vya bure (ofa bora) | Zana | Uoni |
|---|---|---|---|---|---|
| GLM 5.3 Flash | UnoRouter | 1M | 1 maombi/dakika 1 req/min per model | Ndiyo | Ndiyo |
| Gemini 3.8 Flash | Google AI Studio | 1M | Recurring provider quota | Ndiyo | Ndiyo |
| Gemini Flash (latest alias) | Google AI Studio | - | 10 maombi/dakika 20 maombi/siku 250,000 tokeni/dakika ~3M | Ndiyo | Ndiyo |
| Qwen3.8 Flash | ElectronHub | 1M | $0.25/week shared recurring credit | Ndiyo | Ndiyo |
| Gemini 3.7 Flash | Google AI Studio | 1M | 10 maombi/dakika 20 maombi/siku 250,000 tokeni/dakika ~3M | Ndiyo | Ndiyo |
| GPT-5.6 Luna | ElectronHub | 1M | $0.25/week shared recurring credit | Ndiyo | Ndiyo |
Inaonyesha 6 kati ya 67 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.
Tumia embeddings za bure kwa hati zako
Open WebUI inaweza kutengeneza embeddings za hati zako kupitia API inayooana na OpenAI. FreeLLMAPI inatoa /v1/embeddings. Kuhamia kwa mbadala wa embedding hakubadilishi modeli, kwa hivyo vekta zilizo kwenye hifadhi yako zinabaki sambamba.
Mipangilio ya embedding kwa Open WebUI
-e RAG_EMBEDDING_ENGINE=openai \
-e RAG_OPENAI_API_BASE_URL=http://host.docker.internal:3001/v1 \
-e RAG_OPENAI_API_KEY=<unified-key> \
-e RAG_EMBEDDING_MODEL=bge-m3Chagua familia moja ya embedding kisha ibaki nayo. Ukibadilisha familia, tengeneza tena embeddings za hati zote. Modeli za embedding za bure katika katalogi ya leo ni:
| Modeli | Mtoa huduma | Vipimo | Ingizo la juu zaidi | Kikomo cha bure |
|---|---|---|---|---|
Gemini Embeddinggemini-embedding-001 | Google AI Studio | 3072 | 2048 | 100 rpm · 1K req/day |
Nemotron Embed VL 1Bnvidia/llama-nemotron-embed-vl-1b-v2 | NVIDIA NIM | 2048 | 8192 | ~40 rpm |
Nemotron 3 Embed 1B (OR free)nvidia/nemotron-3-embed-1b:free | OpenRouter | 2048 | 32768 | 20 rpm · shared free daily cap |
SEA-LION E5 Embedding 600Maisingapore/SEA-LION-E5-Embedding-600M | SEA-LION | 1024 | 8192 | free · 10 rpm |
SEA-LION ModernBERT Embedding 300Maisingapore/SEA-LION-ModernBERT-Embedding-300M | SEA-LION | 1024 | 8192 | free · 10 rpm |
SEA-LION ModernBERT Embedding 600Maisingapore/SEA-LION-ModernBERT-Embedding-600M | SEA-LION | 1024 | 8192 | free · 10 rpm |
BGE-M3@cf/baai/bge-m3 | Cloudflare Workers AI | 1024 | 8192 | 10K neurons/day (shared) |
EmbeddingGemma 300M@cf/google/embeddinggemma-300m | Cloudflare Workers AI | 768 | 2048 | 10K neurons/day (shared) |
Qwen3 Embedding 0.6B@cf/qwen/qwen3-embedding-0.6b | Cloudflare Workers AI | 1024 | 8192 | 10K neurons/day (shared) |
Cohere Embed v4embed-v4.0 | Cohere | 1536 | 128000 | 1K calls/mo (shared w/ chat) |
Gemini Embedding 2gemini-embedding-2 | Google AI Studio | 3072 | 8192 | free-tier rate limited |
Nemotron 3 Embed 1Bnvidia/nemotron-3-embed-1b | NVIDIA NIM | 2048 | 4096 | ~40 rpm |
PLaMo Embedding 1B (CF)@cf/pfnet/plamo-embedding-1b | Cloudflare Workers AI | 2048 | 4096 | 10K neurons/day (shared) |
BGE Small EN v1.5 (CF)@cf/baai/bge-small-en-v1.5 | Cloudflare Workers AI | 384 | 512 | 10K neurons/day (shared) |
BGE Base EN v1.5 (CF)@cf/baai/bge-base-en-v1.5 | Cloudflare Workers AI | 768 | 512 | 10K neurons/day (shared) |
BGE Large EN v1.5 (CF)@cf/baai/bge-large-en-v1.5 | Cloudflare Workers AI | 1024 | 512 | 10K neurons/day (shared) |
Mistral Embedmistral-embed | Mistral | 1024 | 8192 | free mode · monthly cap (shared) |
Codestral Embedcodestral-embed | Mistral | 1536 | 8192 | free mode · monthly cap (shared) |
Llama Text Embed v2 (Pinecone)llama-text-embed-v2 | Pinecone | 1024 | 2048 | 5M tokens/mo per model |
Multilingual E5 Large (Pinecone)multilingual-e5-large | Pinecone | 1024 | 507 | 5M tokens/mo per model |
Nemotron 3 Embed 1B (Speka)nvidia/nemotron-3-embed-1b | Speka | 2048 | 8192 | $1/month shared across Speka models |
OpenAI: Text Embedding 3 Small (ElectronHub)text-embedding-3-small | ElectronHub | 1536 | 8192 | $0.25/week shared free credits |
OpenAI: Text Embedding 3 Large (ElectronHub)text-embedding-3-large | ElectronHub | 3072 | 8192 | $0.25/week shared free credits |
Nemotron 3 Embed 8B (Inferbase)nvidia-nemotron-3-embed-8b | Inferbase | 4096 | 8192 | 5,000 requests/month; 100 requests and $0.50 usage/day shared |
Qwen 3 Embedding 0.6B (Inferbase)qwen3-embedding-0-6b | Inferbase | 1024 | 8192 | 5,000 requests/month; 100 requests and $0.50 usage/day shared |
Qwen 3 Embedding 4B (Inferbase)qwen3-embedding-4b | Inferbase | 2560 | 8192 | 5,000 requests/month; 100 requests and $0.50 usage/day shared |
Qwen 3 Embedding 8B (Inferbase)qwen3-embedding-8b | Inferbase | 4096 | 8192 | 5,000 requests/month; 100 requests and $0.50 usage/day shared |
LFM2.5 Embedding 350M (LLMTR)liquid/lfm-2.5-embedding-350m-free | LLMTR | 1024 | 512 | Ongoing zero-price route; provider terms apply |
BGE-M3 (SEA-LION)BAAI/bge-m3 | SEA-LION | 1024 | 8192 | free · 10 rpm |
Cohere Embed English v3embed-english-v3.0 | Cohere | 1024 | 512 | 1K calls/mo (shared w/ chat) |
Qwen3 Embedding 8B (OVH)Qwen3-Embedding-8B | OVH AI Endpoints | 4096 | 32768 | free · anonymous 2 req/min per IP |
BGE-M3 (OVH)bge-m3 | OVH AI Endpoints | 1024 | 8192 | free · anonymous 2 req/min per IP |
BGE Multilingual Gemma2 (OVH)bge-multilingual-gemma2 | OVH AI Endpoints | 3584 | 8192 | free · anonymous 2 req/min per IP |
Mistral: Codestral Embed (ElectronHub)codestral-embed | ElectronHub | 1536 | 8192 | $0.25/week shared free credits |
Mistral: Mistral Embed (ElectronHub)mistral-embed | ElectronHub | 1024 | 8192 | $0.25/week shared free credits |
Google: Gemini Embedding 2 (ElectronHub)gemini-embedding-2 | ElectronHub | 3072 | 8192 | $0.25/week shared free credits |
Google: Gemini Embedding 001 (ElectronHub)gemini-embedding-001 | ElectronHub | 3072 | 2048 | $0.25/week shared free credits |
OpenAI: Text Embedding Ada 002 (ElectronHub)text-embedding-ada-002 | ElectronHub | 1536 | 8192 | $0.25/week shared free credits |
Cohere Embed English Light v3embed-english-light-v3.0 | Cohere | 384 | 512 | 1K calls/mo (shared w/ chat) |
Cohere Embed Multilingual v3embed-multilingual-v3.0 | Cohere | 1024 | 512 | 1K calls/mo (shared w/ chat) |
Cohere Embed Multilingual Light v3embed-multilingual-light-v3.0 | Cohere | 384 | 512 | 1K 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.
Utatuzi wa matatizo
| Tatizo | Chanzo | Suluhisho |
|---|---|---|
| Verify Connection inashindwa na hitilafu ya mtandao | localhost inaelekeza kwenye kontena la Open WebUI | Tumia host.docker.internal au mtandao wa pamoja wa Docker, kama katika Hatua ya 1 |
| Verify Connection inashindwa na 401 | Ufunguo si sahihi au una nafasi | Nakili tena ufunguo wa pamoja kutoka ukurasa wa Keys |
| Mamia ya modeli kwenye kichagua | Muunganisho unasoma katalogi yote | Ongeza auto na IDs ulizochagua kwenye Model IDs |
| Modeli moja inashindwa kwa kila ujumbe | Hakuna ufunguo wa mtoa huduma wa modeli hiyo | Ongeza ufunguo wa mtoa huduma, au chagua auto |
| Hitilafu 429 baada ya gumzo nyingi | Funguo zote za watoa huduma kwenye mnyororo zimefikia kikomo chao | Ongeza funguo za watoa huduma zaidi, au subiri muda wa kuweka upya kwenye kichwa cha Retry-After |
| Vigezo vya mazingira havina athari | Thamani zilizohifadhiwa kwenye hifadhidata zinapewa kipaumbele | Hariri muunganisho kwenye Admin Panel |
Ili kuona mtoa huduma aliyejibu gumzo, soma kichwa cha X-Routed-Via cha jibu. Kitazamaji cha Server Logs kwenye menyu ya Analytics ya dashibodi pia kinaonyesha maamuzi ya uelekezaji na muda wa kupumzika wa funguo.
Maswali yanayoulizwa mara kwa mara
Je, kuna API ya bure kwa Open WebUI?
Ndiyo. Open WebUI inafanya kazi na API yoyote inayooana na OpenAI. FreeLLMAPI inachanganya ngazi za bure za watoa huduma nyuma ya URL moja na ufunguo mmoja. Kila ngazi ya bure ina vikomo vyake.
Ninatumia URL gani kwa muunganisho unaooana na OpenAI kwenye Open WebUI?
Tumia URL ya msingi yenye /v1 mwishoni, kwa mfano http://host.docker.internal:3001/v1. Usiongeze /chat/completions. Ikiwa Open WebUI inaendeshwa kwenye Docker, usitumie localhost.
Kwa nini Open WebUI inapuuza mabadiliko yangu ya OPENAI_API_BASE_URL?
Open WebUI inahifadhi mipangilio ya muunganisho kwenye hifadhidata yake baada ya kuanza kwa mara ya kwanza. Thamani zilizo kwenye hifadhidata zinapewa kipaumbele. Hariri muunganisho kwenye Admin Panel, au weka ENABLE_PERSISTENT_CONFIG=False.
Je, Open WebUI inaweza kutumia watoa huduma kadhaa wa bure kwa wakati mmoja?
Ndiyo. Unaweza kuongeza muunganisho mmoja kwa kila mtoa huduma. Kwa FreeLLMAPI, muunganisho mmoja unatosha, kwa sababu kipanga njia kinachagua mtoa huduma na kuhamia kwa anayefuata kikomo kinapozuia ombi.
Modeli ipi ya bure ni bora zaidi kwenye Open WebUI?
Anza na auto. Inafuata mnyororo wako wa akiba. Kwa modeli isiyobadilika, chagua moja yenye dirisha kubwa la muktadha kutoka jedwali la ukurasa huu.
Miongozo inayohusiana
- API ya bure kwa SillyTavern
- Base URL za free API zinazooana na OpenAI
- API ya bure ya embeddings
- API ya LLM bure kwa RAG
- Mipaka ya kiwango cha API ya LLM bure
Vyanzo
- Open WebUI docs: environment variable configuration (imekaguliwa 20 November 2026)
- Open WebUI docs: connect an OpenAI-compatible provider (imekaguliwa 20 November 2026)
- Open WebUI README: Docker installation (imekaguliwa 20 November 2026)
- FreeLLMAPI docs: REST API (routing strategies, model listing, embeddings) (imekaguliwa 20 November 2026)
- FreeLLMAPI docs: Docker deployment (imekaguliwa 20 November 2026)