Powered byLightning AI
← Managed inference

Qwen3.8 27B (image+video)

Coding, long-horizon agents and image/video in one dense model

Start free

Overview

Qwen3.8-27B is the newest generation of Qwen's open-model family and the most capable dense model here: 27B parameters with a native vision encoder, built on the Qwen3.5 architecture. Qwen highlights gains in coding, professional work, research and long-horizon agentic tasks, stronger autonomous planning and handling of environment feedback, and image plus video understanding spanning STEM diagrams, documents and hour-scale video. Context is 262K natively.

Strengths

  • Qwen's strongest coding and agentic dense model
  • Native image and video understanding
  • Thinking can be tuned or turned off per request

Use cases

  • Long-horizon agents that must finish multi-step tasks
  • Coding work that also has to read screenshots or diagrams
  • Document and video analysis

Good to know

Thinking mode is on by default, so responses include a reasoning phase unless you disable it on the request.

API access

One OpenAI-compatible API for every model in the catalog, plus the CLI and the function SDK inside workflows. Sign up free, add a card for $5 in credits, and these requests work:

qwen3.8-27b · curl
curl https://gateway.graphn.ai/v1/chat/completions \
  -H "Authorization: Bearer $GRAPHN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.8-27b",
    "messages": [{"role": "user", "content": "Hello"}]
  }'