Find your model

17 open models behind one OpenAI-compatible API. Search for what you need, and compare prices per million tokens.

Search by name, provider, task, context window or price. Typos are fine.

Check prices via our API

17 models

Copy for agents

Prices in USD per 1M tokens

Model catalogue: what each model is best for, what it accepts, its context window and token prices in US dollars per million tokens
ModelBest forAcceptsContextInputCached inputOutput
GeneralTextImages1M$0.50$0.13$1.50
CodingAgentsText1M$1.40$0.26$4.40
CodingAgentsTextImages1M$2.00$0.20$10.00
FastTextImages1M$0.20$0.05$0.50
CodingAgentsTextImages256K$1.25$0.31$4.50
AgentsText256K$2.50$0.63$6.00
Long documentsTextImages1M$0.40$0.10$2.00
EmbeddingsText32K$0.02$0.00Input only
GeneralTextImages256K$0.40$0.10$2.40
RoutingTextImages1M$0.40$0.10$2.00
RoutingTextImages256K$0.40$0.10$2.40
RoutingTextImages1M$0.40$0.10$2.00
RoutingText1M$1.40$0.26$4.40
FastAgentsText1M$0.25$0.06$0.30
FastGeneralTextImages256K$0.40$0.10$2.40
FastTextImages1M$0.20$0.05$0.50
FastCodingText1M$1.40$0.26$4.40

Retired models (34)

These models are no longer listed. Some old model IDs still work as aliases; others return 404. Use the current model listed alongside each one.

Retired models, the model ID each one had and the model to use instead
Retired modelFormer model IDUse instead
DeepSeek V3.2DeepSeekNot publishedDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
GLM-5Z.aiNot publishedGLM-5.3z-ai/glm-5.3
Kimi K2.5MoonshotNot publishedGLM-5.3 Flashz-ai/glm-5.3-flash
Qwen3.5 397B A17BQwenqwen/qwen3.5-397b-a17bGLM-5.3 Flashz-ai/glm-5.3-flash
Qwen3 235B ThinkingQwenNot publishedDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Qwen3 Next 80B A3B ThinkingQwenqwen/qwen3-next-80b-a3b-thinkingGLM-5.3 Flashz-ai/glm-5.3-flash
Qwen3 32BQwenqwen/qwen3-32bGLM-5.3 Flashz-ai/glm-5.3-flash
Qwen3 30B A3BQwenqwen/qwen3-30b-a3b-instruct-2507DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Qwen3 Coder 30B A3BQwenNot publishedDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Qwen2.5 VL 72BQwenqwen/qwen2.5-vl-72b-instructGLM-5.3 Flashz-ai/glm-5.3-flash
INTELLECT-3PrimeIntellectNot publishedGLM-5.3 Flashz-ai/glm-5.3-flash
Llama 3.3 70BMetameta-llama/llama-3.3-70b-instructDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Gemma 3 27BGooglegoogle/gemma-3-27b-itGLM-5.3 Flashz-ai/glm-5.3-flash
gpt-oss 120BOpenAIopenai/gpt-oss-120bGLM-5.3 Flashz-ai/glm-5.3-flash
Hermes 4 405BNousResearchnousresearch/hermes-4-405bDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Hermes 4 70BNousResearchnousresearch/hermes-4-70bDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Nemotron 3 Ultra 550BNVIDIAnvidia/nemotron-3-ultra-550b-a55bDeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Nemotron Ultra 253BNVIDIAnvidia/llama-3_1-nemotron-ultra-253b-v1DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Nemotron 3 Super 120BNVIDIAnvidia/nemotron-3-super-120b-a12bGLM-5.3 Flashz-ai/glm-5.3-flash
Nemotron 3 Nano 30BNVIDIAnvidia/nvidia-nemotron-3-nano-30b-a3bGLM-5.3 Flashz-ai/glm-5.3-flash
Nemotron 3 Nano OmniNVIDIAnvidia/nemotron-3-nano-omniGLM-5.3 Flashz-ai/glm-5.3-flash
Cosmos 3 Super ReasonerNVIDIAnvidia/cosmos3-super-reasonerGLM-5.3 Flashz-ai/glm-5.3-flash
MiniCPM-V 4.5OpenBMBopenbmb/minicpm-v-4_5GLM-5.3 Flashz-ai/glm-5.3-flash
MiniMax M2.5MiniMaxminimax/minimax-m2.5MiniMax M3minimax/minimax-m3
Kimi K2.6Moonshotmoonshotai/kimi-k2.6Kimi K3.1 Promoonshotai/kimi-k3.1-pro
GLM-5.1Z.aiz-ai/glm-5.1GLM-5.3z-ai/glm-5.3
Qwen3.5 9BQwenqwen/qwen3.5-9bQwen3.8 27Bqwen/qwen3.8-27b
Qwen3 235B A22BQwenqwen/qwen3-235b-a22b-instruct-2507Qwen3.8 2.4T A95Bqwen/qwen3.8-2.4t-a95b
GLM-5.2Z.aiz-ai/glm-5.2GLM-5.3z-ai/glm-5.3
GLM-5.2 InstantZ.aiz-ai/glm-5.2-instantGLM-5.3z-ai/glm-5.3
Kimi K3Moonshotmoonshotai/kimi-k3Kimi K3.1 Promoonshotai/kimi-k3.1-pro
DeepSeek V4 Pro 0813DeepSeekdeepseek/deepseek-v4-pro-0813DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash
Qwen3.8 Flash NextQwenqwen/qwen3.8-flash-nextQwen3.8 27Bqwen/qwen3.8-27b
Qwen3.8 Flash Next InstantQwenqwen/qwen3.8-flash-next-instantQwen3.8 27B Instantqwen/qwen3.8-27b-instant

Pick a model and start building