Base Coming soon
BananaMind-2-Ultra
The next and largest model in the BananaMind 2 family. Announced, not yet released.
BananaMind-2-Ultra is planned as a 400M-parameter decoder-only causal language model with a 16,384-token context window. It has not been released: there are no public weights, no tokenizer, and no benchmark results yet.
- Parameters
- 400M planned
- Context length
- 16,384 tokens
- Status
- Unreleased
Specification
Architecture
- Parameters
- 400M (planned)
- Context length
- 16,384 tokens
- Model type
- Decoder-only causal language model
- Variant
- Base
- Weights
- Not released
- Benchmarks
- None published
Figures are announced targets for an unreleased model. Architecture details, training data, evaluation results, and licensing will be published with the release.
Availability
Not yet released
There are no weights, tokenizer, benchmark results, or license to publish for BananaMind-2-Ultra yet. Released BananaMind models are listed on the models page and on Hugging Face.