120B- input
- $0.00
- output
- $0.00
- Context
- 1Mtokens
- Listed
- 2d ago
- Size
- —parameters
Cost/1M tokens
Language model from Baidu, listed on OpenRouter on November 10, 2026.
- Language model
- Open weights
- 1M context
- Reasoning
- Tools
- Structured output
How it compares
Frequently asked questions
Privacy and data
Yes, ERNIE-5.5-Flash-Preview has open weights, so you can self-host it for full control of your data.
Capabilities and specs
The context window is 1M tokens (1,024,000), and the maximum output is 922K tokens (921,600).
No, input is text only. ERNIE-5.5-Flash-Preview cannot read images, audio, video or PDFs.
Yes, ERNIE-5.5-Flash-Preview supports tool calling (function calling).
Yes, ERNIE-5.5-Flash-Preview supports structured output (JSON schema or JSON mode).
Yes, reasoning is supported and is always on. The reasoning effort levels are high, and the default effort is high.
Quality and benchmarks
Nemotron 3.5 Super 120B A12B by NVIDIA, Kolibri 1 by Aleph Alpha, and Flint by Recursive. They are matched on model kind and release date, with no comparable price.
Nemotron 3.5 Super 120B A12BERNIE-5.5-Flash-Preview vs Nemotron 3.5 Super 120B A12BKolibri 1ERNIE-5.5-Flash-Preview vs Kolibri 1FlintERNIE-5.5-Flash-Preview vs Flint
Licensing and access
ERNIE-5.5-Flash-Preview is open weights.
The ERNIE-5.5-Flash-Preview weights are available on Hugging Face in the baidu/ERNIE-5.5-Flash-Preview repository.
Basics
ERNIE-5.5-Flash-Preview was made by Baidu and was first listed on OpenRouter on November 10, 2026.
The listed facts only state that ERNIE-5.5-Flash-Preview is open weights. No pricing or free-tier information is given.
Generated from Vertical Tabs data, updated November 11, 2026.
Discussion
No comments yet
Be the first to share your thoughts.