Novita AI model deprecations
82 models Novita AI deprecated or shut down, read from its own pages.
Retired 82
- baidu/ernie-4.5-vl-424b-a47b Stopped working
- deepseek/deepseek-v4-flash Stopped working
- deepseek/deepseek-v4-flash-vision-exp Stopped working
- minimax/minimax-m2 Stopped working
- minimax/minimax-m2.1 Stopped working
- minimax/minimax-m2.5-highspeed Stopped working
- mistralai/mistral-nemo Stopped working
- openai/gpt-oss-20b Stopped working
- qwen/qwen-2.5-72b-instruct Stopped working
- qwen/qwen3-235b-a22b-fp8 Stopped working
- qwen/qwen3-235b-a22b-instruct-2507 Stopped working
- qwen/qwen3-235b-a22b-thinking-2507 Stopped working
- qwen/qwen3-coder-30b-a3b-instruct Stopped working
- qwen/qwen3-coder-480b-a35b-instruct Stopped working
- qwen/qwen3-coder-next Stopped working
- qwen/qwen3-embedding-0.6b Stopped working
- qwen/qwen3-max Stopped working
- qwen/qwen3-next-80b-a3b-instruct Stopped working
- qwen/qwen3-omni-30b-a3b-instruct Stopped working
- qwen/qwen3-vl-235b-a22b-instruct Stopped working
- qwen/qwen3-vl-235b-a22b-thinking Stopped working
- qwen/qwen3.5-plus Stopped working
- qwen/qwen3.6-plus Stopped working
- sao10k/l3-8b-lunaris Stopped working
- sao10k/l31-70b-euryale-v2.2 Stopped working
- thudm/glm-4-32b-0414 Stopped working
- zai-org/glm-5-turbo Stopped working
- zai-org/glm-5v-turbo Stopped working
- deepseek/deepseek-r1 Stopped working
- deepseek/deepseek-r1-distill-llama-70b Stopped working
- deepseek/deepseek-v3.1 Stopped working
- deepseek/deepseek-v3.1-terminus Stopped working
- deepseek/deepseek-v3.2 Stopped working
- deepseek/deepseek-v3.2-exp Stopped working
- deepseek/deepseek-v3/community Stopped working
- kwaipilot/kat-coder-pro Stopped working
- inclusionai/ling-3.0-flash-fast Stopped working
- inclusionai/ling-2.6-1t Stopped working
- inclusionai/ling-2.6-flash Stopped working
- inclusionai/ring-2.6-1t Stopped working
- inclusionai/ling-3.0-tiny Stopped working
- qwen/qwen3-14b Stopped working
- qwen/qwen3-30b-a3b-instruct-2507 Stopped working
- qwen/qwen3-30b-a3b-thinking-2507 Stopped working
- qwen/qwen3-8b Stopped working
- qwen/qwen3-next-80b-a3b-thinking Stopped working
- qwen/qwen3-vl-30b-a3b-thinking Stopped working
- qwen/qwen3-vl-32b-instruct Stopped working
- qwen/qwen3-vl-32b-thinking Stopped working
- qwen/qwen3-vl-8b-instruct Stopped working
- qwen/qwen3-vl-8b-thinking Stopped working
- nex-agi/nex-n2-pro Stopped working
- deepseek/deepseek-prover-v2-671b Stopped working
- meta-llama/llama-3-70b-instruct Stopped working
- meta-llama/llama-3-8b-instruct Stopped working
- zai-org/glm-4.5 Stopped working
- xiaomimimo/mimo-v2-flash Stopped working
- baidu/ernie-4.5-21b-a3b Stopped working
- baidu/ernie-4.5-21b-a3b-thinking Stopped working
- baidu/ernie-4.5-300b-a47b-paddle Stopped working
- baidu/ernie-4.5-vl-28b-a3b Stopped working
- meta-llama/llama-3.2-1b-instruct Stopped working
- nousresearch/hermes-2-pro-llama-3-8b Stopped working
- qwen/qwen2.5-32b-instruct Stopped working
- qwen/qwen2.5-7b-instruct Stopped working
- qwen/qwen2.5-vl-72b-instruct Stopped working
- qwen/qwen3-30b-a3b Stopped working
- qwen/qwen3-32b Stopped working
- qwen/qwen3-4b Stopped working
- sao10k/l3-70b-euryale-v2.1 Stopped working
- baichuan/baichuan-m2-32b Stopped working
- deepseek/deepseek-ocr Stopped working
- deepseek/deepseek-r1-0528-qwen3-8b Stopped working
- deepseek/deepseek-r1-distill-qwen-14b Stopped working
- deepseek/deepseek-r1-distill-qwen-32b Stopped working
- google/gemma-3-12b-it Stopped working
- gryphe/mythomax-l2-13b Stopped working
- meta-llama/llama-3.2-3b-instruct Stopped working
- paddlepaddle/paddleocr-vl Stopped working
- qwen/qwen3-8b-fp8 Stopped working
- qwen/qwen3-reranker-8b Stopped working
- thudm/glm-4.1v-9b-thinking Stopped working
Know before your API calls break.
Okapi finds the calls a breaking change hits in your code and opens the pull request that fixes them.
Add to GitHub