Baseten Updates & Release Notes

You can start sending requests to GLM 5.2 today through our Model APIs by calling the OpenAI-compatible endpoint with your Baseten API key. For larger workloads, dedicated deployments are available.

GLM-5.2 is Z.ai's flagship model for agentic engineering is built to perform well on long-horizon coding tasks. GLM-5.2 runs on the Baseten Inference Stack.

curl -X POST https://inference.baseten.co/v1/chat/completions \
-H "Authorization: Api-Key $BASETEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "zai-org/GLM-5.2",
"messages": [{"role": "user", "content": "Refactor this function for readability"}]
}'

For more information and to get started, see our docs.

You can start sending requests to Kimi-K2.7-Code today through our Model APIs by calling the OpenAI-compatible endpoint with your Baseten API key. For larger workloads, dedicated deployments are available.

Kimi K2.7 Code is Moonshot AI's coding-focused model, built for long-horizon agentic engineering tasks with a 262K-token context window. Kimi-K2.7-Code runs on the Baseten Inference Stack.

curl -X POST https://inference.baseten.co/v1/chat/completions \
-H "Authorization: Api-Key $BASETEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "moonshotai/Kimi-K2.7-Code",
  "messages": [{
    "role": "user",
    "content": "Write a binary search in Rust"
  }]
}'

For more information and to get started, see our docs.

Baseten now surfaces engine-native metrics for models served with vLLM or SGLang directly in the Metrics tab.

Baseten automatically detects the engine through your container's code /metrics endpoint, then graphs metrics such as tokens per second, time to first token, KV cache usage, and requests running or queued, no configuration or redeploy required.

vLLM metrics

You can also export these metrics to your own observability stack alongside Baseten's standard metrics.

For more information, see our docs.

You can deploy Nemotron Ultra in one click with our Model APIs. Dedicated deployments are available for larger workloads. NVIDIA Nemotron Ultra, a 550B-parameter mixture-of-experts model with 55B active parameters, is now available through Model APIs with a 202K-token context window.

Call it with the OpenAI or Anthropic SDK, with tool calling, structured outputs, and opt-in reasoning.

curl -X POST https://inference.baseten.co/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Api-Key $BASETEN_API_KEY" \
-d '{
"model": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B",
"messages": [
{
"role": "user",
"content": "Implement Hello World in Python"
}
],
"stream": true,
"stream_options": {
"include_usage": true,
"continuous_usage_stats": true
},
"top_p": 1,
"max_tokens": 1000,
"temperature": 1,
"presence_penalty": 0,
"frequency_penalty": 0
}' \
--no-buffer

For more information, see our docs or get started by talking to us.

The DeepSeek v3.1 and MiniMax M2.5 Model API(s) will be deprecated at 5pm PT on June 17th.

At that time the model ID(s) will become inactive and return an error for all requests. As open source models advance rapidly, we prioritize serving the highest quality models and deprecate models when stronger alternatives are available.

We recommend DeepSeek v4 Pro as an alternative. DeepSeek v4 Pro offers strong agentic and coding capabilities. Just swap in the new Model ID deepseek-ai/DeepSeek-V4-Pro prior to the deprecation date. If you’d like to continue using the previous weights, please contact us about a dedicated deployment of the model.