Title: Gemma 4 31B IT %E2%80%94 gemma-4-31b-it Canonical: https://llmeu.com/lv/models/gemma-4-31b-it/ Markdown mirror: https://llmeu.com/lv/models/gemma-4-31b-it.md Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json Provider: Google DeepMind Licence: apache-2.0 Hosting: catalogue only %E2%80%94 not hosted on LLM EU yet Sources: https://huggingface.co/google/gemma-4-31B-it https://ai.google.dev/gemma/docs/gemma_4_license # Gemma 4 31B IT [Google DeepMind](/lv/providers/google/) P%C4%93d%C4%93jo reizi p%C4%81rskat%C4%ABts: 2026. g. 14. okt. ## Suverenit%C4%81te un mitin%C4%81%C5%A1ana %C5%A0is modelis ir iek%C4%BCauts katalog%C4%81 uzzi%C5%86ai. Tas v%C4%93l nav mitin%C4%81ts LLM EU, un neviens pirm%C4%81s puses galapunkts to %C5%A1eit neapkalpo. Piepras%C4%ABjumi uz to tiek nos%C5%ABt%C4%ABti tie%C5%A1i pakalpojumu sniedz%C4%93jam, saska%C5%86%C4%81 ar %C5%A1%C4%AB pakalpojumu sniedz%C4%93ja noteikumiem. ## Kopsavilkums Gemma 4 31B (gemma-4-31b-it) ir liel%C4%81kais instrukcij%C4%81m piel%C4%81gotais Gemma 4 modelis, ar 31,3 miljardiem parametru saska%C5%86%C4%81 ar Apache-2.0 licenci. Tas apstr%C4%81d%C4%81 tekstu un att%C4%93lus 256K %C5%BEetonu konteksta log%C4%81, un kart%C4%93 ir zi%C5%86ots par 85,2% MMLU Pro, 84,3% GPQA Diamond un 88,4% MMMLU, visi p%C4%81rdev%C4%93ja zi%C5%86oti; t%C4%81 apkalpo%C5%A1anai BF16 re%C5%BE%C4%ABm%C4%81 nepiecie%C5%A1ams liels atmi%C5%86as pa%C4%81trin%C4%81t%C4%81js. ## Specifik%C4%81cijas - Pakalpojumu sniedz%C4%93js Google DeepMind (US) - Konteksta logs 262k - Parametri 31.27B - Modalit%C4%81tes text, vision - Licence apache-2.0 %C2%B7 [licence](https://ai.google.dev/gemma/docs/gemma_4_license) - Atv%C4%93rtie svari yes - Izdo%C5%A1anas datums 2026. g. 11. marts - Uzdevumi [Redz%C4%93%C5%A1ana](/lv/tasks/vision/), [Sprie%C5%A1ana](/lv/tasks/reasoning/), [Daudzvalodu atbalsts](/lv/tasks/multilingual-support/), [Kod%C4%93%C5%A1ana](/lv/tasks/coding/), [Kopsavilk%C5%A1ana](/lv/tasks/summarization/) ## Valodas **Sp%C4%93c%C4%ABga:** en, de, fr, es, it, pt **Adekv%C4%81ta:** nl, pl, sv, cs, ro, el, uk, ru, ar, hi, zh, ja, ko, tr #### Sp%C4%93c%C4%ABga - en - de - fr - es - it - pt #### Adekv%C4%81ta - nl - pl - sv - cs - ro - el - uk - ru - ar - hi - zh - ja - ko - tr Nav verific%C4%93ts: pakalpojumu sniedz%C4%93js nav dokument%C4%93jis %C5%A1o apgalvojumu. %E2%80%94 a language is listed as strong only where a source claims it; the card links the evidence where one exists. ## Kad to izmantot #### Stipr%C4%81s puses - Apache-2.0 licence, unlike earlier Gemma generations - 256K-token context window - text and image input with variable resolution - vendor reports multilingual coverage in over 140 languages #### Ierobe%C5%BEojumi - all published scores are self-reported by the vendor - 31B parameters need a large-memory accelerator at BF16 - no evaluation has been run by LLM EU ## Bie%C5%BE%C4%81k uzdotie jaut%C4%81jumi ### Is Gemma 4 31B IT served on LLM EU? It has no endpoint in the catalogue, so it is catalogue only %E2%80%94 not hosted on LLM EU yet, with no region and no retention to report. A card whose status is "listed" is a catalogue entry, not a running deployment. ### Under what licence is Gemma 4 31B IT released? Gemma 4 31B IT is recorded under apache-2.0. The recorded licence text is https://ai.google.dev/gemma/docs/gemma_4_license. The weights are downloadable (open_weights is true). LLM EU republishes the licence id and the link and gives no legal advice; the licence text governs. ### How large is Gemma 4 31B IT, and what context window does it have? Gemma 4 31B IT is recorded at 31.3B (safetensors total 31,273,088,876) (source: https://huggingface.co/api/models/google/gemma-4-31B-it), with a context window of 262144 tokens (source: https://huggingface.co/google/gemma-4-31B-it). ### Cenas %C5%A0is modelis ir iek%C4%BCauts katalog%C4%81 uzzi%C5%86ai. Tas v%C4%93l nav mitin%C4%81ts LLM EU, un neviens pirm%C4%81s puses galapunkts to %C5%A1eit neapkalpo. Piepras%C4%ABjumi uz to tiek nos%C5%ABt%C4%ABti tie%C5%A1i pakalpojumu sniedz%C4%93jam, saska%C5%86%C4%81 ar %C5%A1%C4%AB pakalpojumu sniedz%C4%93ja noteikumiem. ### Verifik%C4%81cijas statuss Ja laukam nav avota, kart%C4%ABt%C4%93 tas tiek par%C4%81d%C4%ABts k%C4%81 nep%C4%81rbaud%C4%ABts, nevis aizpild%C4%ABts. Tuk%C5%A1s lauks ir inform%C4%81cija. LLM EU Verified noz%C4%ABm%C4%93 dokument%C4%93ts, nevis juridiski sertific%C4%93ts. Tas nav AI akta atbilst%C4%ABbas nov%C4%93rt%C4%93jums, un LLM EU nav ES strukt%C5%ABra. ### Versijas un varianti - catalogue %C2%B7 Tikai katalog%C4%81 %C2%B7 [weights](https://huggingface.co/google/gemma-4-31B-it) ### Piepras%C4%ABjuma piem%C4%93rs ``` curl https://api.llmeu.com/v1/chat/completions \ -H "Authorization: Bearer $LLMEU_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"gemma-4-31b-it","messages":[{"role":"user","content":"Hello"}]}' ``` %C5%A0im modelim nav izvietots rota%C4%BCu laukums: neviens galapunkts to neapkalpo, t%C4%81p%C4%93c piepras%C4%ABjumam neb%C5%ABtu kurp doties. ### Etaloni - MMLU Pro: **85.2% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-31B-it) - GPQA Diamond: **84.3% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-31B-it) - AIME 2026, no tools: **89.2% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-31B-it) - LiveCodeBench v6: **80.0% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-31B-it) - MMMLU (multilingual MMLU): **88.4% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-31B-it) ### Avoti - [huggingface.co](https://huggingface.co/google/gemma-4-31B-it) - [ai.google.dev](https://ai.google.dev/gemma/docs/gemma_4_license) ### Kartes izmai%C5%86u %C5%BEurn%C4%81ls - 2026. g. 11. marts %E2%80%94 Izdo%C5%A1anas datums - 2026. g. 14. okt. %E2%80%94 P%C4%93d%C4%93jo reizi p%C4%81rskat%C4%ABts P%C4%93d%C4%93jo reizi p%C4%81rskat%C4%ABts: 2026. g. 14. okt. ### Uzdevumi - [Redz%C4%93%C5%A1ana](/lv/tasks/vision/) - [Kopsavilk%C5%A1ana](/lv/tasks/summarization/)