Qwen3.8 2.4T A95B

Overview. Qwen3.8 2.4T A95B is an open-weight Qwen3.8 reasoning model built as a mixture-of-experts architecture with 2.4T total parameters and 95B active parameters. It requires thinking for every interaction, making it suited to developers building text-based applications that benefit from deliberate reasoning and a 262,144-token context window.

Key strengths. Artificial Analysis reports an Intelligence Index score of 39.9 for the model. Its 262,144-token context window supports long text inputs, while the MoE design routes each token through 10 of 512 experts and activates 95B parameters per forward pass.

When to use it. Use this for text-based reasoning tasks when long-context handling and deliberate inference matter. It is also a fit for applications that can accommodate required thinking on every interaction and do not need native vision or audio input.

Limitations. Thinking cannot be disabled, so the model is not suited to workflows that require a non-reasoning mode. It accepts text input and produces text output only; the available findings do not establish tool calling, structured output, streaming behavior, or a training-data cutoff.

Sovereignty note. Qwen is the creator, and the available findings do not establish the creator's country or CLOUD Act exposure. Endpoint data residency and training-on-input policy are not specified in the available findings, so no Sovereign classification can be made.