Occiglot 7B EU5
Europe's open polyglot LLM — 7B covering English, German, French, Spanish, Italian.
About
Occiglot is a non-profit academic research collective (DFKI, hessian.AI and partners) building open LLMs for European linguistic diversity and AI sovereignty. Its v0.1 release (April 2024) continued-pretrained Mistral-7B on 700B multilingual tokens plus ~1B instruction tokens, producing per-language and combined EU5 models for English, German, French, Spanish, and Italian — all Apache 2.0 on Hugging Face, with an Ollama build available.
The roadmap targets all 24 official EU languages plus regional ones, backed by a 1T-token non-English corpus; later releases added Llama-3-based German models with 32K context via the DiscoResearch collaboration.
Honest limits: v0.1 models are explicitly intermediary research checkpoints, not safety-aligned (may produce problematic outputs); 7B scale and 2024 vintage mean they trail modern multilingual models; per-language quality varies; the ambitious 24-language goal is still in progress.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 6/10
Scored 6/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 55; Trending score 45/100 from researcher assessment.