DeepSeek R1 🔥 Top Reasoner DeepSeek 671B MoE 128K Context
DeepSeek's breakthrough open-weights reasoning model with long chain-of-thought and stellar math/coding benchmark scores.
7 Free Providers Available
View Providers ▼ SiliconFlow Direct CN 🟢 203ms
deepseek-ai/DeepSeek-R1 Copy ID
🎁 Free ¥16 credit voucher upon verification (180d validity) (Covers ~1,600 tokens · No card required)
Cloudflare Workers AI Global Route 🟢 240ms
@cf/deepseek-ai/deepseek-r1-distill-qwen-32b Copy ID
🎁 10,000 Neurons free compute daily (Global 300+ edge nodes · Daily reset)
NVIDIA NIM Global Route 🟢 274ms
deepseek-ai/deepseek-r1 Copy ID
🎁 Free 1,000 trial credits (NVIDIA enterprise container · 40 RPM)
Together AI Global Route 🟢 280ms
deepseek-ai/DeepSeek-R1 Copy ID
🎁 Free $5 trial bonus on signup (Low-latency distributed inference cluster)
Novita AI Global Route 🟢 281ms
deepseek/deepseek-r1 Copy ID
🎁 新用户注册赠送体验金 ($1 ~ $50) (64k Context · 试用期内有效 (90天))
GitHub Models Global Route 🟢 304ms
DeepSeek-R1 Copy ID
🎁 150 free calls daily for GitHub accounts (Azure high-availability hosting · Free dev tier)
SambaNova Cloud Global Route 🟢 304ms
DeepSeek-R1 Copy ID
🎁 200,000 free tokens daily (SN40L hardware acceleration · No card)
DeepSeek V3 / V4-Flash ⚡ Versatile Workhorse DeepSeek 671B MoE 1M Context
DeepSeek's flagship foundational general model with MLA architecture, extreme efficiency, and 1M context.
6 Free Providers Available
View Providers ▼ ModelScope (阿里) Direct CN 🟢 152ms
deepseek-ai/DeepSeek-V4-Flash-0731 Copy ID
🎁 2,000 req/day开源推理调用 (64k Context · 每日自动重置)
AMD Radeon Cloud Direct CN 🟢 157ms
DeepSeek-V4-Flash-0731 Copy ID
🎁 $10 free compute daily (Resets daily) (~10M–100M tokens · 1M context · Direct line)
SiliconFlow Direct CN 🟢 175ms
deepseek-ai/DeepSeek-V3 Copy ID
🎁 Free ¥16 credit voucher upon verification (180d validity) (Covers ~16M tokens · Fast direct line)
Vercel AI Gateway Global Route 🟢 205ms
deepseek-v3.2-thinking Copy ID
🎁 $5/mo free team credit (Monthly reset) (GitHub 1-click login · Global Anycast edge)
NVIDIA Global Route 🟢 294ms
deepseek-ai/deepseek-v4.1-flash Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
Community Open Fleet Global Route 🟢 150ms
deepseek-chat Copy ID
🎁 Daily 定时重置Free配额 (128k Context · 每日自动重置)
Llama 3.3 70B 👑 Open Standard Meta 70B Dense 128K Context
Meta's flagship 70B dense model matching 405B capabilities with industry-wide deployment and ecosystem support.
7 Free Providers Available
View Providers ▼ Cerebras Cloud Global Route 🟢 127ms
llama-3.3-70b Copy ID
🎁 $5 wafer-scale compute bonus on signup (CS-3 wafer chip · 1,800+ tok/s throughput)
SambaNova Cloud Global Route 🟢 206ms
Meta-Llama-3.3-70B-Instruct Copy ID
🎁 Daily 200,000 Tokens Free Quota (SN40L chip · Daily auto-refresh · No card)
Cloudflare Workers AI Global Route 🟢 258ms
@cf/meta/llama-3.3-70b-instruct-fp8-fast Copy ID
🎁 10,000 Neurons free compute daily (Global CDN edge routing · Daily auto-refill)
Together AI Global Route 🟢 265ms
meta-llama/Llama-3.3-70B-Instruct-Turbo Copy ID
🎁 Free $5 credit bonus on signup (Turbo ultra-fast inference cluster)
NVIDIA NIM Global Route 🟢 266ms
meta/llama-3.3-70b-instruct Copy ID
🎁 Free 1,000 trial credits (Official high concurrency container · 40 RPM)
Hugging Face Global Route 🟢 275ms
meta-llama/Llama-3.3-70B-Instruct Copy ID
🎁 Daily 定时重置Free配额 (128k Context · 每日自动重置)
GitHub Models Global Route 🟢 295ms
Meta-Llama-3.3-70B-Instruct Copy ID
🎁 150 free calls daily for GitHub accounts (No card required · Azure high availability)
Qwen 2.5 72B 🇨🇳 Multilingual Flagship Alibaba 72B Dense 128K Context
Alibaba Qwen's premier open weights model, dominant in multilingual comprehension, math, and structured data.
4 Free Providers Available
View Providers ▼ Alibaba Bailian Direct CN 🟢 115ms
qwen-plus Copy ID
🎁 1M token free trial pack on signup (Official enterprise gateway · Alibaba backbone)
ModelScope (Alibaba) Direct CN 🟢 130ms
Qwen/Qwen2.5-72B-Instruct Copy ID
🎁 2,000 free requests daily (Resets daily) (Alibaba official open source community)
SambaNova Cloud Global Route 🟢 202ms
Qwen2.5-72B-Instruct Copy ID
🎁 每分钟 20 次请求超高并发 (128k Context · 每日自动重置)
Together AI Global Route 🟢 280ms
Qwen/Qwen2.5-72B-Instruct-Turbo Copy ID
🎁 $5 free credit on signup (Global low-latency delivery)
Qwen 3.8 / Flash-Next ⚡ Hyper-Speed LPU Alibaba 27B / Flash 131K Context
Alibaba Qwen's latest fast architecture running at hyper-speed on Groq LPU and AMD compute fabrics.
5 Free Providers Available
View Providers ▼ AMD Radeon Cloud Direct CN 🟢 135ms
Qwen3.8-Flash-Next Copy ID
🎁 $10 free compute daily (Resets daily) (AMD GPU cluster · Ultra-fast direct line)
Groq Global Route 🟢 146ms
qwen/qwen3.8-27b Copy ID
🎁 14,400 calls daily ultra-fast (30 RPM) (Groq LPU wafer chip · 300+ tok/s · No card)
ModelScope (Alibaba) Direct CN 🟢 161ms
Qwen/Qwen3.8-Flash-Next Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
Vercel AI Gateway Global Route 🟢 205ms
qwen-3.8-flash-next Copy ID
🎁 $5/mo free team credit (Monthly reset) (GitHub OAuth access · No card required)
DeepInfra Global Route 🟢 286ms
Qwen/Qwen3.8-Flash Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
Gemini 2.5 Flash 🌟 1M Massive Context Google Multimodal Native 1M Context
Google's native multimodal Flash model with 1M context, accepting text, image, audio, and video inputs natively.
4 Free Providers Available
View Providers ▼ Vercel AI Gateway Global Route 🟢 211ms
gemini-2.5-flash Copy ID
🎁 $5/mo free team credit (Monthly refresh) (Vercel global Anycast edge network)
Google AI Studio Global Route 🟢 211ms
gemini-2.5-flash Copy ID
🎁 1,500 calls daily 100% free (15 RPM) (Google official free quota · 1M context · Email login)
DeepInfra Global Route 🟢 271ms
google/gemini-3.1-flash-lite Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
OpenRouter Free Zone Global Route 🟢 340ms
google/gemini-2.0-flash-exp:free Copy ID
🎁 Official 0-cost endpoint (:free) (Multi-provider aggregation & fallback)
GLM-4-Flash / 5.3 🇨🇳 Permanent Free & High Concurrency Zhipu AI (GLM) High Concurrency 128K Context
Zhipu AI's officially permanent free model with massive concurrency, ideal for high-volume automated workflows.
6 Free Providers Available
View Providers ▼ Zhipu AI (GLM) Direct CN 🟢 159ms
glm-4v-flash Copy ID
🎁 Daily 定时重置Free配额 (128k Context · 每日自动重置)
ModelScope (Alibaba) Direct CN 🟢 166ms
ZhipuAI/GLM-4.7-Flash Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
Vercel AI Gateway Global Route 🟢 205ms
glm-5.3-flash Copy ID
🎁 $5/mo free team credit (Monthly refresh) (统一 API Gateway 接入)
DeepInfra Global Route 🟢 279ms
zai-org/GLM-5.3-Flash Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
NVIDIA Global Route 🟢 294ms
z-ai/glm-5.3-flash Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
Zhipu AI (GLM) Direct CN 🟢 150ms
glm-4-flash Copy ID
🎁 Official permanent free tier · 0 voucher deducted (128K context · Direct BGP line)
GPT-4o mini 🌍 Dev Classic Standard OpenAI Compact Multimodal 128K Context
OpenAI's compact flagship workhorse with exceptional instruction-following, speed, and cost efficiency.
3 Free Providers Available
View Providers ▼ GitHub Models Global Route 🟢 147ms
gpt-4o-mini Copy ID
🎁 150 free calls daily for GitHub accounts (Microsoft Azure native hosting · No card)
LLM7.io Global Route 🟢 260ms
gpt-4o-mini Copy ID
🎁 10 req/min no account (40 with key) (Zero signup barrier · Instant call)
Pollinations AI Global Route 🟢 312ms
openai/gpt-4o-mini Copy ID
🎁 Daily 定时重置Free配额 (128k Context · 每日自动重置)
GPT-OSS 120B 🚀 Huge Open Weights OpenAI 120B Open 131K Context
OpenAI's 120B open architecture running at incredible token speeds on Groq LPU wafer chips.
2 Free Providers Available
View Providers ▼ Groq Global Route 🟢 92ms
openai/gpt-oss-120b Copy ID
🎁 1,000 calls daily (30 RPM) (LPU hyper-speed output · No card required)
Cerebras Cloud Global Route 🟢 108ms
gpt-oss-120b Copy ID
🎁 注册送 $5 晶圆算力体验金 (Wafer-scale hardware acceleration)
Codestral 22B / Mistral 💻 Coding & FIM Specialist Mistral AI 22B Coding Special 256K Context
Mistral AI's 22B code-centric model tuned for fill-in-the-middle (FIM) and multi-file code editing across 80+ languages.
2 Free Providers Available
View Providers ▼ NVIDIA NIM Global Route 🟢 260ms
mistralai/codestral-22b-instruct-v0.1 Copy ID
🎁 Free 1,000 点数试用额度 (40 RPM enterprise concurrency)
Mistral AI Global Route 🟢 329ms
codestral-latest Copy ID
🎁 Free tier mode & La Plateforme (256K code context · Simple email signup)
Kimi K2.6 / K3 🇨🇳 Long-Context Pioneer Moonshot AI (Kimi) Long-form MoE 128K Context
Moonshot AI's flagship long-context MoE model with outstanding document synthesis and tool-use capabilities.
2 Free Providers Available
View Providers ▼ Moonshot AI (Kimi) Direct CN 🟢 163ms
moonshot-v1-8k Copy ID
🎁 Free ¥15 trial bonus on signup (Official native platform · Low latency BGP)
NVIDIA NIM Global Route 🟢 260ms
moonshotai/kimi-k2.6 Copy ID
🎁 1,000 Credits 试用点数 (NVIDIA NIM 官方托管实例)
MiniMax Text-01 🇨🇳 4M Huge Context MiniMax 456B MoE (45.9B active) 1M Context
MiniMax's high-capacity MoE architecture supporting massive multi-million context windows with superb creative coherence.
4 Free Providers Available
View Providers ▼ Groq Global Route 🟢 88ms
minimaxai/minimax-m2.7 Copy ID
🎁 1,000 calls daily ultra-fast (Groq 芯片加速)
ModelScope (Alibaba) Direct CN 🟢 168ms
MiniMax/MiniMax-M1-80k Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
MiniMax Direct CN 🟢 171ms
MiniMax-Text-01 Copy ID
🎁 15M free tokens bonus on signup (Valid for all foundation models · Direct line)
SambaNova Cloud Global Route 🟢 202ms
MiniMax-M3 Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
Nemotron 3 Super / Ultra 🛡️ Agents & Tool Use NVIDIA 120B ~ 550B MoE 1M Context
NVIDIA's flagship open models tuned for autonomous agents, high-precision function calling, and safety.
2 Free Providers Available
View Providers ▼ Requesty Global Route 🟢 228ms
nvidia/nemotron-3-super-120b-a12b Copy ID
🎁 200 free requests daily (Resets daily) (1M context window · $0 permanent free tier)
NVIDIA NIM Global Route 🟢 571ms
nvidia/llama-3.1-nemotron-70b-instruct Copy ID
🎁 1,000 Credits 试用点数 (NVIDIA 官方优化运行时容器)
Doubao 1.5 Pro 🇨🇳 ByteDance Flagship VolcEngine (ByteDance) ByteDance Native 128K Context
ByteDance's flagship model powering millions of consumers, stellar in native Chinese nuances, dialogue, and search.
1 Free Providers Available
View Providers ▼ Volcengine (ByteDance Doubao) Direct CN 🟢 125ms
doubao-1.5-pro-32k Copy ID
🎁 500,000 free tokens bonus (Domestic BGP direct backbone · Ultra-low latency)
ERNIE Speed / 4.0 🇨🇳 Official Permanent Free Baidu AI Cloud Permanent Free 128K Context
Baidu Qianfan's permanently free model for developers, zero cost with native Chinese knowledge base retrieval.
3 Free Providers Available
View Providers ▼ ModelScope (Alibaba) Direct CN 🟢 144ms
PaddlePaddle/ERNIE-4.5-300B-A47B-PT Copy ID
🎁 官方 0 元专区 · Daily 自动补充 (32k Context · 每日自动重置)
百度千帆 Direct CN 🟢 174ms
ernie-lite-8k Copy ID
🎁 官方Permanent FreeFree Calls (8k Context · 永久有效)
Baidu Qianfan Direct CN 🟢 174ms
ernie-speed-128k Copy ID
🎁 Official permanent free tier · Unlimited calls (Baidu backbone · Permanent free tier)
Tencent Hunyuan Lite 🇨🇳 256K Permanent Free Tencent Cloud 256K Permanent Free 256K Context
Tencent Hunyuan's permanently free model with 256K context, optimized for enterprise WeChat and Tencent Cloud workflows.
1 Free Providers Available
View Providers ▼ Tencent Hunyuan Direct CN 🟢 138ms
hunyuan-lite Copy ID
🎁 Official permanent free · 256K context (Tencent Cloud backbone · Permanent free)
SenseTime SenseNova 6.8 🇨🇳 Daily Login Bonus Credits SenseTime Flash Lite / SenseChat 128K Context
SenseTime's flagship reasoning foundation with daily login reward points for abundant ongoing free developer usage.
1 Free Providers Available
View Providers ▼ SenseTime SenseNova Direct CN 🟢 169ms
sensenova-6.8-flash-lite Copy ID
🎁 1M points on signup + Daily check-in bonus (Direct dedicated line · Low response latency)
iFlytek Spark Lite 🇨🇳 Voice & NLP Specialist iFlytek Permanent Free 128K Context
iFlytek's permanently free model leveraging decades of speech synthesis and NLP expertise.
1 Free Providers Available
View Providers ▼ iFlytek Spark Direct CN 🟢 175ms
lite Copy ID
🎁 Official permanent free · 2M token bonus (Direct cloud line · Superior NLP capabilities)
