
inclusionAI/Ling-3.0-flash-int4
The INT4 variant of an efficient 124B hybrid MoE model with 5.1B active parameters, 256K context, and strong coding and agentic reasoning.
inclusionAI’s Ling-3.0-flash-int4 is an INT4-quantized version of its 124B-parameter Ling-3.0-flash reasoning model, built on a hybrid-linear sparse MoE architecture that activates only 5.1B parameters per token. With a 256K context window and near-BF16 benchmark performance, it is designed for efficient production inference across coding, deep research, instruction following, and long-running agentic workflows.
| Attribute | Value |
|---|---|
| Parameter Size | 124B-A5.1B |
| Context Length | 256k |
| Input Modalities | Text |
| Output Modalities | Text |
| Supported Features | Tool Use JSON mode Structured Outputs Reasoning |
Input
export BASETEN_MODEL_ID="<your-model-id>"
curl -N -X POST \
"https://model-${BASETEN_MODEL_ID}.api.baseten.co/environments/production/sync/v1/chat/completions" \
-H "Authorization: Bearer $BASETEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "inclusionAI/Ling-3.0-flash-int4",
"messages": [{"role": "user", "content": "hello!"}],
"chat_template_kwargs": {"enable_thinking": true},
"stream": true,
"temperature": 0.6,
"top_k": 20,
"top_p": 0.95
}'