Models
24 model families · 16 providers · sorted by best free quota
GPT-4o / GPT-4.1
OpenAIOpenAI flagship series
GPT-5
OpenAIOpenAI GPT-5 series
GPT-OSS
OpenAIOpenAI open-weight models
C Cloudflare Workers AI
@cf/openai/gpt-oss-120b · +1
∞/d
H Hugging Face Inference
openai/gpt-oss-120b
∞/d
N NVIDIA NIM
openai/gpt-oss-120b · +1
∞/d 40/min
O Ollama Cloud
gpt-oss:120b · +1
∞/d
S SambaNova Cloud
gpt-oss-120b
∞/d 20/min
C Cerebras
gpt-oss-120b
14k/d 30/min
G Groq
openai/gpt-oss-120b · +2
1k/d 30/min
Gemini
GoogleGemini 2.5 / 3 series
Gemma 3
GoogleGoogle open Gemma models
Llama 3.x
MetaMeta Llama 3 series
C Cloudflare Workers AI
@cf/meta/llama-3.3-70b-instruct-fp8-fast · +3
∞/d
H Hugging Face Inference
meta-llama/Llama-3.3-70B-Instruct
∞/d
N NVIDIA NIM
meta/llama-3.3-70b-instruct · +4
∞/d 40/min
S SambaNova Cloud
Meta-Llama-3.3-70B-Instruct
∞/d 20/min
S SiliconFlow
meta-llama/Meta-Llama-3.1-8B-Instruct
∞/d 1000/min
G Groq
llama-3.1-8b-instant · +1
14k/d 30/min
C Cerebras
llama3.1-8b
14k/d 30/min
G GitHub Models
meta/llama-3.3-70b-instruct · +1
50/d 10/min
Llama 4
MetaMeta Llama 4 series
C Cloudflare Workers AI
@cf/meta/llama-4-scout-17b-16e-instruct
∞/d
N NVIDIA NIM
meta/llama-4-maverick-17b-128e-instruct
∞/d 40/min
S SambaNova Cloud
Llama-4-Maverick-17B-128E-Instruct
∞/d 10/min
G Groq
meta-llama/llama-4-scout-17b-16e-instruct
1k/d 30/min
G GitHub Models
meta/llama-4-scout-17b-16e-instruct · +1
50/d 10/min
DeepSeek V3
DeepSeekDeepSeek V3 chat / code
DeepSeek R1
DeepSeekDeepSeek R1 reasoning
Qwen 3
AlibabaAlibaba Qwen 3 family
C Cloudflare Workers AI
@cf/deepseek-ai/deepseek-r1-distill-qwen-32b
∞/d
H Hugging Face Inference
Qwen/Qwen3-235B-A22B
∞/d
N NVIDIA NIM
qwen/qwen3-coder-480b-a35b-instruct
∞/d 40/min
O Ollama Cloud
qwen3-coder:480b · +4
∞/d
S SiliconFlow
Qwen/Qwen3-8B
∞/d 1000/min
C Cerebras
qwen-3-235b-a22b-instruct-2507
14k/d 30/min
M ModelScope
Qwen/Qwen3-235B-A22B-Instruct-2507 · +5
2k/d
G Groq
qwen/qwen3-32b
1k/d 60/min
Kimi K2
MoonshotMoonshot Kimi K2
GLM
ZhipuZhipu GLM series
B BigModel
glm-4.7-flash · +3
∞/d 30/min
N NVIDIA NIM
z-ai/glm-5.1
∞/d 40/min
O Ollama Cloud
glm-5.1 · +3
∞/d
S SiliconFlow
THUDM/GLM-4-9B-0414 · +2
∞/d 1000/min
M ModelScope
ZhipuAI/GLM-4.6 · +1
2k/d
A AIHubMix
coding-glm-5.1-free · +7
500/d 5/min
C Cerebras
zai-glm-4.7
100/d 10/min
O OpenRouter
z-ai/glm-4.5-air:free
50/d 20/min
MiniMax M2
MiniMaxMiniMax M2 series
Mistral / Magistral / Pixtral
MistralMistral conversational + vision
Codestral
MistralMistral coding model
Phi-4
MicrosoftMicrosoft Phi models
Command
CohereCohere Command family
Aya
CohereCohere multilingual Aya
Nemotron
NVIDIANVIDIA Nemotron series
Step
StepFunStepFun Step series
MiMo
XiaomiXiaomi MiMo series
Ling
AntAnt Group Ling series
ALLaM
SDAIAArabic ALLaM
Groq Compound
GroqGroq Compound system
Other
20 models
Models not matched to a known family. Add a family to model-families.ts to categorize them.
deepseek-ai/deepseek-v4-pro deepseek-ai/deepseek-v4-flash
openai/o4-mini openai/o3 openai/o3-mini openai/o1 xai/grok-3 xai/grok-3-mini microsoft/mai-ds-r1 ai21-labs/ai21-jamba-1.5-large
Qwen/Qwen2.5-7B-Instruct Qwen/Qwen2.5-Coder-7B-Instruct Qwen/Qwen2.5-VL-7B-Instruct tencent/Hunyuan-MT-7B
@cf/qwen/qwen2.5-coder-32b-instruct @cf/qwen/qwq-32b
deepseek-v4-pro deepseek-v4-flash devstral-2:123b cogito-2.1:671b