Title: Phi-4-multimodal-instruct | LLM EU Canonical: https://llmeu.com/hr/models/phi-4-multimodal-instruct/ Markdown mirror: https://llmeu.com/hr/models/phi-4-multimodal-instruct.md Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json Provider: Microsoft Licence: mit Hosting: catalogue only %E2%80%94 not hosted on LLM EU yet Sources: https://huggingface.co/microsoft/Phi-4-multimodal-instruct https://arxiv.org/abs/2503.01743 # Phi-4-multimodal-instruct [Microsoft](/hr/providers/microsoft/) Posljednji pregled: 14. lis 2026. ## Suverenost i hosting Ovaj model je u katalogu za referencu. Jo%C5%A1 nije hostan na LLM EU i nijedan endpoint prvog lica ga ovdje ne poslu%C5%BEuje. Zahtjevi za njega idu izravno pru%C5%BEatelju, pod uvjetima tog pru%C5%BEatelja. ## Sa%C5%BEetak Phi-4-multimodal-instruct je model od 5.6B koji prihva%C4%87a tekst, slike i govor u jednoj kontrolnoj to%C4%8Dki, s kontekstom od 128K tokena i MIT licencom. Kartica modela izvje%C5%A1tava 55.1 na MMMU, 93.2 na DocVQA i 62.4 na MathVista, te navodi podr%C5%BEane jezike zasebno za svaki ulazni modalitet. To je kompaktna opcija za dokumentne i govorne cjevovode koji bi ina%C4%8De zahtijevali dva modela. ## Specifikacije - Pru%C5%BEatelj Microsoft (US) - Kontekstni prozor 131k - Parametri 5.57B - Modaliteti text, vision, audio - Licenca mit %C2%B7 [licence](https://opensource.org/license/mit) - Otvoreni utezi yes - Datum izdanja 24. velj 2025. - Zadaci [Vizija](/hr/tasks/vision/), [Sa%C5%BEimanje](/hr/tasks/summarization/), [Vi%C5%A1ejezi%C4%8Dna podr%C5%A1ka](/hr/tasks/multilingual-support/), [Niska cijena](/hr/tasks/cheap/) ## Jezici **Jako:** en, de, fr, es, it, pt **Zadovoljavaju%C4%87e:** nl, pl, sv, da, fi, cs, ru, uk, ar, zh, ja, ko, hi, tr #### Jako - en - de - fr - es - it - pt #### Zadovoljavaju%C4%87e - nl - pl - sv - da - fi - cs - ru - uk - ar - zh - ja - ko - hi - tr Nije provjereno: pru%C5%BEatelj nije dokumentirao ovu tvrdnju. %E2%80%94 a language is listed as strong only where a source claims it; the card links the evidence where one exists. ## Kada ga koristiti #### Prednosti - text, image and speech input at 5.6B parameters - 128K-token context for document batches - MIT licence - 23 declared languages including most large EU languages #### Ograni%C4%8Denja - modality-specific quality is below dedicated single-modality models - English-centric training data - no evaluation has been run by LLM EU ## %C4%8Cesto postavljana pitanja ### Is Phi-4-multimodal-instruct 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 Phi-4-multimodal-instruct released? Phi-4-multimodal-instruct is recorded under mit. The recorded licence text is https://opensource.org/license/mit. 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 Phi-4-multimodal-instruct, and what context window does it have? Phi-4-multimodal-instruct is recorded at 5.57B (safetensors total 5,574,460,384) (source: https://huggingface.co/api/models/microsoft/Phi-4-multimodal-instruct), with a context window of 131072 tokens (source: https://huggingface.co/microsoft/Phi-4-multimodal-instruct). ### Cijene Ovaj model je u katalogu za referencu. Jo%C5%A1 nije hostan na LLM EU i nijedan endpoint prvog lica ga ovdje ne poslu%C5%BEuje. Zahtjevi za njega idu izravno pru%C5%BEatelju, pod uvjetima tog pru%C5%BEatelja. ### Status provjere Kada polje nema izvor, kartica ga prikazuje kao neprovjereno umjesto da ga popuni. Prazno polje je informacija. LLM EU Verified zna%C4%8Di dokumentirano, ne pravno certificirano. To nije procjena sukladnosti s AI Actom i LLM EU nije tijelo EU-a. ### Verzije i varijante - catalogue %C2%B7 Samo u katalogu %C2%B7 [weights](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) ### Primjer zahtjeva ``` curl https://api.llmeu.com/v1/chat/completions \ -H "Authorization: Bearer $LLMEU_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"phi-4-multimodal-instruct","messages":[{"role":"user","content":"Hello"}]}' ``` Za ovaj model nije postavljen playground: nijedan endpoint ga ne poslu%C5%BEuje, pa zahtjev ne bi imao kamo oti%C4%87i. ### Benchmarkovi - MMMU: **55.1 (vendor-reported)** [source](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) - DocVQA: **93.2 (vendor-reported)** [source](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) - MathVista (testmini): **62.4 (vendor-reported)** [source](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) ### Izvori - [huggingface.co](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) - [arxiv.org](https://arxiv.org/abs/2503.01743) ### Povijest izmjena kartice - 24. velj 2025. %E2%80%94 Datum izdanja - 14. lis 2026. %E2%80%94 Posljednji pregled Posljednji pregled: 14. lis 2026. ### Zadaci - [Vizija](/hr/tasks/vision/)