Flower Labs, the company behind the Flower federated learning framework, released Endeavor 1.1 on November 3, its third model this year after the UK sovereign 7B model Lizzy in April and Endeavor 1.0 in September. The update adds image input, an optional EU-only GPU inference tier, and Flower-managed, private-cloud or on-premises deployment, and promises readable thinking traces plus ownership of outputs that customers may use to train their own models. Vendor-reported scores are GPQA 94.4, HumanEval 99.9, AIME 2026 99.9 and MMMU-Pro 81.1, but the comparison uses GPT-5.6 Sol and Claude Fable 5 rather than either lab's newest models, and access is request-only for now.
Key Takeaways
- ✓Vendor-reported: GPQA rose from 92.0 in 1.0 to 94.4 (about 30% fewer errors), HumanEval from 98.2 to 99.9; IFEval is 94.5 and MMMU-Pro 81.1.
- ✓In the vendor table, GPT-5.6 Sol scores GPQA 94.1, IFEval 95.9 and MMMU-Pro 83.0, and Claude Fable 5 scores GPQA 92.6; Endeavor does not lead on IFEval or MMMU-Pro.
- ✓New EU-only GPU inference with SLAs, with UK, Middle East, Asia and Africa regions planned.
- ✓The pitch is control: raw thinking traces, ownership of outputs and traces, and the right to fine-tune or train your own models on them.
- ✓Flower says the model builds on mature open-weight models via continual pre-training, targeted post-training and model integration, without naming the base or parameter count; access is request-only.
Key Decision Metrics at a Glance
Not enough VRAM? Compare cloud API and self-hosting costs
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Background and Pain Points
With closed frontier APIs, the vendor controls availability, terms and model versions. OpenAI exposes only thinking summaries, Anthropic usually withholds raw traces, and both restrict training competing models on outputs. European buyers also face inference-location limits: Flower notes that Azure's UK-only per-token offering still covers only GPT-4o and GPT-4.1 mini. Flower Labs, known for its federated learning framework, aims Endeavor at buyers who want frontier capability under their own control.
Architecture and Mechanism
Flower says Endeavor 1.1 builds on mature open-weight models, adds UK-specialist knowledge from Lizzy, and combines them through continual pre-training, targeted post-training and model integration, without disclosing the base model or parameter count. The team stresses the whole system over the weights: it tunes inference-time reasoning, context handling, tool use and failure recovery across several in-house coding and agent harnesses, and uses FlowerBench to run real enterprise workflows inside participating organizations' own environments without moving their data.
Benchmarks
All figures are vendor-reported: GPQA 94.4, HumanEval 99.9, IFEval 94.5, AIME 2026 99.9 and MMMU-Pro 81.1. In the same table GPT-5.6 Sol scores 94.1, 95.1, 95.9, 99.9 and 83.0, and Claude Fable 5 scores 92.6, 97.0, 91.7, 99.9 and 81.2. Three caveats: the comparison skips both labs' newest flagships, HumanEval and AIME are near saturation, and there is no independent evaluation or SWE-bench-style agentic coding result yet.
Getting Started
Access is by request at flower.ai/models/endeavor, with Flower-managed inference (global or EU GPUs), private cloud or on-premises deployment, plus enterprise SLAs. It fits teams that need data residency, auditable reasoning, or want to distill specialist models from their own traces. Test it on your real tasks against your current model before relying on the composite score.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.