Is Endeavor 1.1 state of the art?Pending benchmarks.Announced Nov 3, 2026; no independent results yet.
Flower Labs' language model with readable thinking traces; version 1.1 adds image input alongside text. Served by request through Flower's Responses-style API, with EU GPU inference options and private deployment; no public weights.
The details so far
- Lab
- Flower Labs
- Announced
- Nov 3, 2026
- API
- Waitlist · By request on Flower's API (api.flower.ai/v1, Responses API) as flwrlabs/endeavor-1.1; works with Codex and OpenCode
- Open weights
- Closed · Private distribution only, with availability and formats confirmed during onboarding
- Type
- Language model with readable thinking traces
- New in 1.1
- Image input alongside text
- Context
- Up to 1M tokens (Flower's supplied Codex configuration uses 128K)
- Deployment
- EU GPU inference options, plus private or on-premises deployment
- Parameters
- Not disclosed
- License
- Not disclosed
- Earlier versions
- Endeavor 1.0 (still available by request) and Lizzy 7B
Tags marked unsure are not stated on the lab's own pages.
What would let us call it
Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.
- Publicly announced with a source
- Usable by the public (API or weights)
- Listed on an independent board we track
- Enough results for a rank in a category
- Holds across two snapshots (crowns only)
What the lab claims
Self-reported. Shown for context; not used in any verdict.
Lab claim: GPQA 94.4, HumanEval 99.9, IFEval 94.5, AIME 2026 99.9 and MMMU-Pro 81.1.
Lab claim: from Endeavor 1.0 to 1.1, GPQA rises from 92.0 to 94.4 (about 30% fewer errors) and HumanEval from 98.2 to 99.9.
Lab claim: the highest composite score (equal-weighted GPQA, AIME 2026, HumanEval and IFEval) of the OpenAI and Anthropic models in Flower's comparison.
Sources
LABFlower Labs announcement on Xx.com/flwrlabs/status/2117523422456807552 · Nov 3, 2026LABIntroducing Endeavor 1.1 (Flower blog)flower.ai/blog/2026-11-03-introducing-endeavor-1.1 · Nov 3, 2026LABEndeavor model page (flower.ai)flower.ai/models/endeavorLABFlower docs: Endeavorflower.ai/docs/model/endeavor.htmlTracking ends Feb 1, 2027 unless the API or weights become available.