New blocks

DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the DeepSeek V3 model and performs really well on a variety of tasks.

CUTIEHOOKD

CUTIEHOOK

Voyage's next-generation embedding model optimized for code retrieval

Voyage's generalist reranker optimized for quality with multilingual support

Mistral Codestral 2501, 260k

Gemini Text Embedding 004

Qwen2.5-72B-Instruct, 32k

Llama-3.1-Nemotron-70B-Instruct, 131k

DeepSeek-R1-Distill-Qwen-32B, 131k

DeepSeek-R1-Distill-Llama-70B, 131k

DeepSeek V3 0324, 160k CTX

Qwen/QwQ-32B, 128k CTX

Llama 4 Scout DeepInfra, 320k CTX

Llama 4 Maverick

The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing

hiu

test

참고문서

월드 기본 문서

api관련

문서