Data as of November 17, 2026 · How the score is built
Olala mini 7A1B
Decision readingOlala mini 7A1B is tracked, but not publicly ranked yet. The profile exposes 0 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Email me when its price, rank or availability changes.
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Mini
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not publishedOVH: OVH announces Olala mini 7A1B
- Context window
- 32KOVH: OVH announces Olala mini 7A1B
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- OVH announced Olala mini 7A1B on November 17, 2026, and will launch it officially at the OVHcloud AI Summit on November 19. OVH says the weights, the pre-trained base model, a technical report and AI Act compliance documentation will be published on Hugging Face under Apache 2.0; no Hugging Face repository was public when BenchLM checked on November 17. OVH says Qwant already uses the model for its flash answers. No hosted API or token price has been published.
- Cloud regions
- Not tracked yet
- Lifecycle
- Coming Soon
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordOVH: OVH announces Olala mini 7A1B
- Self-host
- Weight availability is pending
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Olala mini 7A1B, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.
Olala mini 7A1B has no verified publication yet for weight access. Its 32K context window is documented separately from that pending weight status. The profile records its reasoning mode as unspecified.
OVH announced Olala mini 7A1B on November 17, 2026, and will launch it officially at the OVHcloud AI Summit on November 19. OVH says the weights, the pre-trained base model, a technical report and AI Act compliance documentation will be published on Hugging Face under Apache 2.0; no Hugging Face repository was public when BenchLM checked on November 17. OVH says Qwant already uses the model for its flash answers. No hosted API or token price has been published.
OVH's first model in its Olala family, built from scratch by its OVHai LLM lab. OVH describes a mixture-of-experts model with 7B total and 1B active parameters spread across 256 experts per layer, hybrid attention that combines Differential-TPA and Mamba-3, BF16 weights and a 32,000-token context window. OVH says it runs on a GPU with as little as 8 GB of memory and trained it on three European public supercomputers (Jupiter, MeluXina and Jean Zay). The announcement publishes no benchmark results, so BenchLM stores no scored rows and the model stays unranked. OVH says version 1.1, due by the end of 2026, will raise the context window to 128,000 tokens.
The profile has no source-displayable benchmark row yet.
Last updated November 17, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does Olala mini 7A1B perform overall in AI benchmarks?
Olala mini 7A1B does not have any source-displayable benchmark rows yet, so this profile does not assign a public score or rank. Documented specifications remain visible, while score-led charts and claims stay unavailable until a published evaluation can be attached to the exact model.
Does Olala mini 7A1B have full benchmark coverage on BenchLM?
No. Olala mini 7A1B currently has 0 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Olala mini 7A1B?
Olala mini 7A1B has a documented context window of 32K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.