Baidu's efficient long-context model, built on open GLM weights with Baidu Search's Lightning research and Baidu Sparse Attention.
Tencent's Hunyuan 4 release, a substantial upgrade over Hunyuan 4 Preview for long-horizon productivity and agentic work.
No shared demos for these models yet.
No captured outputs for these models yet.
Both ERNIE-Light-DT-1.0 and Hunyuan 4 are capable AI models. The best choice depends on your specific use case: consider pricing, context window, speed, and which capabilities (vision, reasoning, audio) you need.
Hunyuan 4 has a lower blended cost. ERNIE-Light-DT-1.0: $1.13 input / $3.94 output. Hunyuan 4: $0.42 input / $1.25 output.
Hunyuan 4 has a larger context window: ERNIE-Light-DT-1.0 supports 1M tokens vs Hunyuan 4 at 1M tokens.
Hunyuan 4 supports vision/image input, but ERNIE-Light-DT-1.0 does not.
Key differences: Hunyuan 4 is significantly cheaper; only Hunyuan 4 supports vision input. Compare full specs on this page.
If cost is your priority, choose the cheaper option. If you need the highest intelligence for complex tasks, pick the higher-scoring model. For long documents or codebases, choose the larger context window. You can try both ERNIE-Light-DT-1.0 and Hunyuan 4 for free on idapt.app to see which performs better for your specific needs.