{ "schemaVersion": 1, "benchmark": "baa", "name": { "sk": "BAA benchmark LLM modelov", "en": "BAA LLM benchmark" }, "url": { "sk": "https://dubovskylabs.com/baa-benchmark/", "en": "https://dubovskylabs.com/en/baa-benchmark/" }, "dateModified": "2026-08-08", "creator": { "name": "Marek Dubovsky", "url": "https://dubovskylabs.com/" }, "methodology": { "sk": "Rovnaký prompt; reálne spustenie výstupu a manuálne praktické hodnotenie hrateľnosti a chýb.", "en": "The same prompt; each output was launched and manually evaluated for practical playability and bugs." }, "prompt": { "originalLanguage": "sk", "original": "Chcel by som vytvorit jednoduchu arkadovu hru, trocha akoby to bola nintendo hra ale v 3D s pohladu z tretej osoby ale trocha vzdialena kamera. Hra bude v prostredi luk v horach na kopcovitom vlnenom terene kde sa pasu ovce, Vacsina plochy budu luky, v pozadi hory a lesy a nejake domy. Hrac bude hrat za ovciarskeho alebo ineho psa ktory sa pouziva pri zahanani oviec, nech si vie vybrat z par plemien ktore sa budu lisit niektorymi vlastnostami. Hrac bude mat moznost sa pohybovat sipkami, medzerou alebo nejakym vhodnym key na klavesnici bude moct brechat co bude moct pouzit aj na ovce teda ak to realne pomaha pri zahanani, ale aj na odplasenie inych zvierat, napr. inych psov, vlkov, medveda, ludi okrem pastiera ktory sa bude tiez nachadzat na luke. Ovce nech reaguju na psa cize budu sa pohybovat v stade ale budu sa pohybovat tak ako ich pes zazenie, nech beekaju (baa), ale nech to je arkadove. Podstata hry ma byt zaviest ovce do kosiara cize, z bodu A do bodu B a celit nastraham ako su obcasne vypady inych psov, vynimocne vlkov, alebo medveda. Vlci a medved nech su nieco ako bosovia alebo proste vacsie prekazky v dalsich levelov. Vymysli si mensie lahsie prekazky v prvych leveloch. Levelov nech je nejaky rozumny pocet, mysli na to ze toto ma byt kratka hra na pobavenie na par minut max par desiatok minut mozno 1 hodina ak by sa to hralo do konca do posledneho levela. Na mape by sa mohli objavovat aj ine zvierata, navhrni napr. kravy, lisky obcas niekde nejake vtaky atd. proste priroda. Nech su okrem luk aj nejaka ta polna cesta, obcas nejake prekazdky ci uz v podobe ohrad, utesov, krikov cez ktore ovce neprejdu. Grafika nech je pekna, farebna, arkadova, trocha nech pripomina akoby to bolo nintendo hra ale nekopiruj ziadnu konkretnu hru respektive jej dizajn. Hru napis v html/javascript/", "englishTranslation": "I would like to create a simple arcade game that feels a little like a Nintendo game, but in 3D, from a third-person perspective with a somewhat distant camera. The game should be set in mountain meadows, on hilly, rolling terrain where sheep graze. Most of the area should consist of meadows, with mountains, forests, and a few houses in the background. The player should control a sheepdog or another type of dog used for herding sheep, with a choice of several breeds that differ in some of their traits. The player should be able to move with the arrow keys and bark using the space bar or another suitable keyboard key. Barking should also affect the sheep, if that realistically helps with herding, and should scare away other animals—for example other dogs, wolves, and a bear—as well as people, except for the shepherd, who should also be present in the meadow. The sheep should react to the dog: they should move as a flock, while also moving in response to the way the dog herds them. They should bleat (“baa”), and the game should have an arcade feel. The main objective should be to guide the sheep into a pen, from point A to point B, while facing hazards such as occasional attacks by other dogs and, more rarely, wolves or a bear. Wolves and the bear should act as bosses, or simply as larger obstacles in later levels. Come up with smaller, easier obstacles for the early levels. There should be a reasonable number of levels; keep in mind that this is meant to be a short game for fun, lasting a few minutes, at most a few dozen minutes, or perhaps one hour if played all the way through the final level. Other animals could also appear on the map; include, for example, cows, foxes, and the occasional bird—in short, nature. In addition to the meadows, include a dirt road and occasional obstacles such as fences, cliffs, and bushes that the sheep cannot pass through. The graphics should be attractive, colorful, and arcade-like, somewhat reminiscent of a Nintendo game, but do not copy any specific game or its design. Write the game in HTML/JavaScript." }, "provenancePolicy": { "sk": "Dátum, verzia promptu a typ behu sú uvedené iba vtedy, keď boli zaznamenané; historické medzery sa neodhadujú.", "en": "A run date, prompt version, and run type are included only when recorded; historical gaps are not estimated." }, "runs": [ { "rank": 1, "model": "Claude Opus 5", "publisher": "Anthropic", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Najzábavnejšia a najprepracovanejšia hra", "bugs": "Bez zistených chýb", "note": "" }, "en": { "assignment": "Matched the assignment", "playability": "Most fun and most polished game", "bugs": "No issues found", "note": "" } }, "description": { "sk": "Najlepší a najkompletnejší výsledok benchmarku: osem levelov, štyri plemená psov, prepracované správanie stáda, dravce, gamepad aj natívne dotykové ovládanie.", "en": "The benchmark's best and most complete result: eight levels, four dog breeds, detailed flock behavior, predators, gamepad support, and native touch controls." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/claude-opus5.png", "demo": "https://dubovskylabs.com/projects/baa/claude-opus-5/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 2, "model": "Claude Fable 5", "publisher": "Anthropic", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Plne hrateľný", "bugs": "Minimum, pár drobných chýb", "note": "Reasoning effort 4/6, nie maximum.\nDrobná oprava viewportu na Macoch." }, "en": { "assignment": "Matched the assignment", "playability": "Fully playable", "bugs": "Minimal, a few minor issues", "note": "Reasoning effort 4/6, not maximum.\nSmall Mac viewport fix." } }, "description": { "sk": "Jeden z najlepších výstupov benchmarku: splnil zadanie, je plne hrateľný a má minimum drobných chýb.", "en": "One of the benchmark's best outputs: matched the assignment, is fully playable, and has only minor issues." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/claude-fable5.png", "demo": "https://dubovskylabs.com/projects/baa/claude-fable-5/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 3, "model": "Kimi K3", "publisher": "Moonshot AI", "modelId": "kimi-k3", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "2,8T", "en": "2.8T" }, "requirements": { "sk": [ { "kind": "weights", "text": "1,56 TB" }, { "kind": "context", "text": "≈147,9 GB / 100K" } ], "en": [ { "kind": "weights", "text": "1.56 TB" }, { "kind": "context", "text": "≈147.9 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Plne hrateľný", "bugs": "Minimum, drobné chyby", "note": "Cena generovania 7,31 USD." }, "en": { "assignment": "Matched the assignment", "playability": "Fully playable", "bugs": "Minimal, minor issues", "note": "Generation cost: USD 7.31." } }, "description": { "sk": "Jeden z najlepších výstupov: veľmi pekná grafika, zábavná hrateľnosť a kvalitné správanie NPC postáv.", "en": "One of the best outputs: strong visuals, fun gameplay, and high-quality NPC behavior." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/kimi-k3.png", "demo": "https://dubovskylabs.com/projects/baa/kimi-k3/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "moonshotai/Kimi-K3@9f62e4e9fffbd0a83ddd60e1c209d828994b3569", "checkedAt": "2026-08-07", "sources": [ { "id": "kimi-k3-model", "kind": "official checkpoint", "url": "https://huggingface.co/moonshotai/Kimi-K3/tree/9f62e4e9fffbd0a83ddd60e1c209d828994b3569", "revision": "9f62e4e9fffbd0a83ddd60e1c209d828994b3569", "accessedAt": "2026-08-07" }, { "id": "kimi-k3-fla-reference", "kind": "reference runtime source", "url": "https://github.com/fla-org/flash-linear-attention/tree/613210444e1e18b2e4f28a0dad36ca007df8ae48", "revision": "613210444e1e18b2e4f28a0dad36ca007df8ae48 (FLA 0.5.2; audit-selected pin)", "accessedAt": "2026-08-07" } ] } }, { "rank": 4, "model": "GPT-5.6 Sol", "publisher": "OpenAI", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Hrateľný", "bugs": "Málo chýb", "note": "" }, "en": { "assignment": "Matched the assignment", "playability": "Playable", "bugs": "Few issues", "note": "" } }, "description": { "sk": "Kompletný, vizuálne čistý výstup so slovenským UI, výberom plemena, šiestimi levelmi a funkčným stádom.", "en": "Complete, visually clean output with Slovak UI, breed selection, six levels, and a working flock." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gpt5-6-sol.png", "demo": "https://dubovskylabs.com/projects/baa/gpt-5.6-sol/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 5, "model": "Claude Opus 4.8", "publisher": "Anthropic", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Hrateľný", "bugs": "Málo chýb", "note": "" }, "en": { "assignment": "Matched the assignment", "playability": "Playable", "bugs": "Few issues", "note": "" } }, "description": { "sk": "Hrateľný cloudový výstup: splnil zadanie a má len málo chýb.", "en": "Playable cloud output: matched the assignment and has only a few issues." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/claude-opus.png", "demo": "https://dubovskylabs.com/projects/baa/claude-opus-4.8/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 6, "model": "Grok 4.5", "publisher": "xAI", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Hrateľný", "bugs": "Bez zjavnej vážnej chyby", "note": "" }, "en": { "assignment": "Matched the assignment", "playability": "Playable", "bugs": "No obvious serious issue", "note": "" } }, "description": { "sk": "Hrateľný 3D výstup s prehľadnou krajinou, stádom, minimapou a viacstupňovým herným menu.", "en": "Playable 3D output with a readable landscape, flock, minimap, and a multi-step game menu." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/grok-4-5.png", "demo": "https://dubovskylabs.com/projects/baa/grok-4.5/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 7, "model": "Gemini 3.5 Flash", "publisher": "Alphabet", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Hrateľný", "bugs": "Málo chýb", "note": "Vyžadoval 1 opravný prompt." }, "en": { "assignment": "Matched the assignment", "playability": "Playable", "bugs": "Few issues", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Takmer na úrovni Opusu: pekná arkádová grafika, funkčné stádo a hrateľný výsledok.", "en": "Almost on par with Opus: clean arcade visuals, a working flock, and a playable result." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gemini-3-5-flash.png", "demo": "https://dubovskylabs.com/projects/baa/gemini-3.5-flash/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 8, "model": "Gemini 3.6 Flash", "publisher": "Alphabet", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Hrateľný", "bugs": "Bez vážnych chýb", "note": "" }, "en": { "assignment": "Matched the assignment", "playability": "Playable", "bugs": "No serious bugs", "note": "" } }, "description": { "sk": "Hrateľný výsledok bez vážnych chýb, ale s prázdnym svetom a minimom detailov.", "en": "A playable result without serious bugs, but with an empty world and very little detail." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gemini-3-6-flash.png", "demo": "https://dubovskylabs.com/projects/baa/gemini-3.6-flash/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 9, "model": "Qwen 3.6 27B ThinkingCap", "publisher": "Alibaba", "modelId": "thinkingcap-qwen3-6-27b", "runtime": { "sk": "Otvorené váhy, GGUF", "en": "Open-weight, GGUF" }, "parameters": { "sk": "27B", "en": "27B" }, "requirements": { "sk": [ { "kind": "weights", "text": "16,8 GB" }, { "kind": "context", "text": "≈6,71 GB / 100K" } ], "en": [ { "kind": "weights", "text": "16.8 GB" }, { "kind": "context", "text": "≈6.71 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Hrateľný", "bugs": "Bez vážnych chýb", "note": "ThinkingCap úprava od Bottlecapu. Generovanie trvalo 17 minút; bez opravného promptu." }, "en": { "assignment": "Partially matched the assignment", "playability": "Playable", "bugs": "No serious bugs", "note": "ThinkingCap modification by Bottlecap. Generated in 17 minutes; no corrective prompt." } }, "description": { "sk": "Prekvapivo vydarená lokálna verzia: spustila sa na prvý pokus, bola hneď hrateľná a nemala vážne chyby.", "en": "A surprisingly successful local version: it started on the first try, was immediately playable, and had no serious bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen3-6-27b-thinkingcap.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.6-27b-thinkingcap/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "bottlecapai/ThinkingCap-Qwen3.6-27B-GGUF@0982db1be5e1e6cf7163ba89d7f63a9f18d2f4f0", "checkedAt": "2026-08-07", "sources": [ { "id": "thinkingcap-gguf", "kind": "publisher checkpoint", "url": "https://huggingface.co/bottlecapai/ThinkingCap-Qwen3.6-27B-GGUF/tree/0982db1be5e1e6cf7163ba89d7f63a9f18d2f4f0", "revision": "0982db1be5e1e6cf7163ba89d7f63a9f18d2f4f0", "accessedAt": "2026-08-07" }, { "id": "qwen3-6-27b-config", "kind": "official config", "url": "https://huggingface.co/Qwen/Qwen3.6-27B/blob/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9/config.json", "revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 10, "model": "Muse Spark 1.2", "publisher": "Meta", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Hrateľný", "bugs": "Veľa chýb, po čase zamŕza", "note": "Zaznamenaný čas: 4 min 19 s." }, "en": { "assignment": "Partially matched the assignment", "playability": "Playable", "bugs": "Many bugs; freezes after a while", "note": "Recorded time: 4 min 19 sec." } }, "description": { "sk": "Ambiciózna 3D hra so šiestimi levelmi, ktorá je hrateľná, no obsahuje veľa chýb a po čase zamŕza.", "en": "An ambitious six-level 3D game that is playable, but contains many bugs and freezes after a while." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/muse-spark-1-2.png", "demo": "https://dubovskylabs.com/projects/baa/muse-spark-1.2/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 11, "model": "GPT-5.6 Terra", "publisher": "OpenAI", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie", "playability": "Hrateľný", "bugs": "Veľa chýb", "note": "Zaznamenaný čas: 20 min." }, "en": { "assignment": "Matched the assignment", "playability": "Playable", "bugs": "Many bugs", "note": "Recorded time: 20 min." } }, "description": { "sk": "Rozsiahla farebná 3D hra so šiestimi levelmi, tromi plemenami a natívnym dotykovým ovládaním; je hrateľná, no obsahuje veľa chýb.", "en": "An extensive, colorful 3D game with six levels, three breeds, and native touch controls; it is playable but contains many bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gpt-5-6-terra.png", "demo": "https://dubovskylabs.com/benchmarks/baa/apps/gpt-5-6-terra/index.html", "provenance": { "generatedAt": "2026-08-08", "testedAt": "2026-08-08", "promptVersion": "baa-original-sk", "runType": "one-shot" }, "modelSource": null }, { "rank": 12, "model": "Qwen 3.5 397B A17B", "publisher": "Alibaba", "modelId": "qwen3-5-397b-a17b", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "397B", "en": "397B" }, "requirements": { "sk": [ { "kind": "weights", "text": "235,7 GB" }, { "kind": "context", "text": "≈3,27 GB / 100K" } ], "en": [ { "kind": "weights", "text": "235.7 GB" }, { "kind": "context", "text": "≈3.27 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Hrateľný", "bugs": "Veľa vážnych chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Playable", "bugs": "Many serious bugs", "note": "" } }, "description": { "sk": "Hra sa dá spustiť a ovládať, no obsahuje veľa vážnych chýb a horšie čitateľnú scénu.", "en": "The game starts and can be controlled, but it contains many serious bugs and a less readable scene." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen3-5-397b-a17b.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.5-397b-a17b/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "Qwen/Qwen3.5-397B-A17B-GPTQ-Int4@df333de344e06b492063d1ca04e2d4ee878f7595", "checkedAt": "2026-08-07", "sources": [ { "id": "qwen3-5-397b-config", "kind": "official config", "url": "https://huggingface.co/Qwen/Qwen3.5-397B-A17B/blob/8472618112abcbd45acbcdc58436aff4233c23f7/config.json", "revision": "8472618112abcbd45acbcdc58436aff4233c23f7", "accessedAt": "2026-08-07" }, { "id": "qwen3-5-397b-int4", "kind": "official 4-bit checkpoint", "url": "https://huggingface.co/Qwen/Qwen3.5-397B-A17B-GPTQ-Int4/tree/df333de344e06b492063d1ca04e2d4ee878f7595", "revision": "df333de344e06b492063d1ca04e2d4ee878f7595", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 13, "model": "MiniMax M2.7", "publisher": "MiniMax", "modelId": "minimax-m2-7", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "229,9B", "en": "229.9B" }, "requirements": { "sk": [ { "kind": "weights", "text": "≈139,9 GB" }, { "kind": "context", "text": "≈25,4 GB / 100K" } ], "en": [ { "kind": "weights", "text": "≈139.9 GB" }, { "kind": "context", "text": "≈25.4 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Hrateľný", "bugs": "Veľa vážnych chýb", "note": "Postavy sa takmer vôbec nehýbu." }, "en": { "assignment": "Partially matched the assignment", "playability": "Playable", "bugs": "Many serious bugs", "note": "Characters barely move." } }, "description": { "sk": "Vizuálne zrozumiteľný výstup, ktorý sa dá ovládať, ale postavy sa takmer nehýbu a hra má veľa vážnych chýb.", "en": "Visually readable output that accepts controls, but characters barely move and the game has many serious bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/minimax-m2-7.png", "demo": "https://dubovskylabs.com/projects/baa/minimax-m2.7/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "MiniMaxAI/MiniMax-M2.7@d494266a4affc0d2995ba1fa35c8481cbd84294b", "checkedAt": "2026-08-07", "sources": [ { "id": "minimax-m2-7-model", "kind": "official checkpoint", "url": "https://huggingface.co/MiniMaxAI/MiniMax-M2.7/tree/d494266a4affc0d2995ba1fa35c8481cbd84294b", "revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b", "accessedAt": "2026-08-07" }, { "id": "minimax-m2-7-config", "kind": "official config", "url": "https://huggingface.co/MiniMaxAI/MiniMax-M2.7/blob/d494266a4affc0d2995ba1fa35c8481cbd84294b/config.json", "revision": "d494266a4affc0d2995ba1fa35c8481cbd84294b", "accessedAt": "2026-08-07" }, { "id": "minimax-m2-7-nvfp4", "kind": "vendor 4-bit conversion", "url": "https://huggingface.co/nvidia/MiniMax-M2.7-NVFP4/tree/e79701cb1f9dce8fe5395b9ed2b20170beebecde", "revision": "e79701cb1f9dce8fe5395b9ed2b20170beebecde", "accessedAt": "2026-08-07" }, { "id": "llama-cpp-q4-k-m", "kind": "runtime documentation", "url": "https://github.com/ggml-org/llama.cpp/blob/15586e2d7165570fb3aa7c26e0d442e289ef69de/tools/quantize/README.md", "revision": "15586e2d7165570fb3aa7c26e0d442e289ef69de", "accessedAt": "2026-08-07" } ] } }, { "rank": 14, "model": "Qwen 3.8 Max Preview", "publisher": "Alibaba", "modelId": "qwen3-8-max-preview", "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": { "sk": "2,4T", "en": "2.4T" }, "requirements": { "sk": [], "en": [] }, "evaluation": { "sk": { "assignment": "Zadanie splnil vo veľkej miere", "playability": "Hrateľný", "bugs": "Zvláštne ovládanie, málo ďalších chýb", "note": "Generovanie sa z neznámeho dôvodu zastavilo pred dokončením." }, "en": { "assignment": "Fulfilled most of the assignment", "playability": "Playable", "bugs": "Unusual controls, few other issues", "note": "Generation stopped unexpectedly before completion." } }, "description": { "sk": "Rozsiahly 3D výstup so štyrmi plemenami a ôsmimi levelmi; je hrateľný, ale ťažkopádny pre zvláštne ovládanie. Generovanie sa navyše nečakane zastavilo.", "en": "An extensive 3D output with four breeds and eight levels; it is playable but cumbersome because of its unusual controls. Generation also stopped unexpectedly." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen-3-8-max.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.8-max/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "qwen3.8-max-preview (hosted endpoint observed 2026-08-07)", "checkedAt": "2026-08-07", "sources": [ { "id": "qwen3-8-endpoint", "kind": "official hosted-model documentation", "url": "https://www.alibabacloud.com/help/en/model-studio/token-plan-harness-tool", "revision": "documentation updated 2026-07-19", "accessedAt": "2026-08-07" }, { "id": "qwen3-8-announcement", "kind": "official model announcement", "url": "https://x.com/Alibaba_Qwen/status/2078759124914098291", "revision": "status 2078759124914098291 published 2026-07-19", "accessedAt": "2026-08-07" } ] } }, { "rank": 15, "model": "DeepSeek V4 Flash 0731 DS4", "publisher": "DeepSeek", "modelId": "deepseek-v4-flash-0731", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "284B", "en": "284B" }, "requirements": { "sk": [ { "kind": "weights", "text": "164,6 GB" }, { "kind": "context", "text": "≈1,22 GB / 100K" } ], "en": [ { "kind": "weights", "text": "164.6 GB" }, { "kind": "context", "text": "≈1.22 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Hrateľný", "bugs": "Zvláštna kamera a množstvo chýb", "note": "Použitá úspornejšia Q2 kvantizácia, ktorá sa zmestila do 96 GB RAM. Čas: 1 h 49 min." }, "en": { "assignment": "Partially matched the assignment", "playability": "Playable", "bugs": "Unusual camera and many bugs", "note": "A more memory-efficient Q2 quantization was used, fitting within 96 GB of RAM. Time: 1 h 49 min." } }, "description": { "sk": "Rozsiahla osemlevelová 3D hra so štyrmi plemenami; je hrateľná, ale ťažkopádna pre zvláštnu kameru a obsahuje množstvo chýb.", "en": "An extensive eight-level 3D game with four breeds; it is playable but cumbersome because of its unusual camera and contains many bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/deepseek-v4-flash-0731.png", "demo": "https://dubovskylabs.com/projects/baa/deepseek-v4-flash-0731-ds-4-q2/sheepdog.html", "provenance": { "generatedAt": "2026-08-08", "testedAt": "2026-08-08", "promptVersion": "baa-original-sk", "runType": "one-shot" }, "modelSource": { "modelRevision": "deepseek-ai/DeepSeek-V4-Flash-0731@7872f01b1d1fe23eabc4c98b48bffcef5a386062", "checkedAt": "2026-08-08", "sources": [ { "id": "deepseek-v4-flash-0731-model", "kind": "official checkpoint and inference code", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731/tree/7872f01b1d1fe23eabc4c98b48bffcef5a386062", "revision": "7872f01b1d1fe23eabc4c98b48bffcef5a386062", "accessedAt": "2026-08-08" }, { "id": "deepseek-v4-flash-core", "kind": "official architecture figures", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash/tree/60d8d70770c6776ff598c94bb586a859a38244f1", "revision": "60d8d70770c6776ff598c94bb586a859a38244f1", "accessedAt": "2026-08-08" }, { "id": "deepseek-v4-flash-0731-ds4-q4", "kind": "pinned DS4 Q4 GGUF artifact manifest", "url": "https://huggingface.co/antirez/deepseek-v4-gguf/blob/e7f04037032990db0346398d249baf9fb9df1ccc/DeepSeek-V4-Flash-Q4KExperts-F16HC-F16Compressor-F16Indexer-Q8Attn-Q8Shared-Q8Out-chat-v2-imatrix-0731.gguf", "revision": "e7f04037032990db0346398d249baf9fb9df1ccc", "accessedAt": "2026-08-08" }, { "id": "ds4-downloader", "kind": "pinned DS4 artifact mapping", "url": "https://github.com/antirez/ds4/blob/b0309611041655f4e45671cfd9c9886aff161406/download_model.sh", "revision": "b0309611041655f4e45671cfd9c9886aff161406", "accessedAt": "2026-08-08" }, { "id": "ds4-runtime", "kind": "pinned DS4 cache allocator source", "url": "https://github.com/antirez/ds4/blob/b0309611041655f4e45671cfd9c9886aff161406/ds4.c", "revision": "b0309611041655f4e45671cfd9c9886aff161406", "accessedAt": "2026-08-08" } ] } }, { "rank": 16, "model": "GPT-5.5", "publisher": "OpenAI", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Vizuálne pekný, ale prakticky nehrateľný výstup s množstvom chýb.", "en": "Visually strong, but practically unplayable output with many bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gpt5-5.png", "demo": "https://dubovskylabs.com/projects/baa/gpt-5.5/dist/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 17, "model": "MiniMax M3", "publisher": "MiniMax", "modelId": "minimax-m3", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "428B", "en": "428B" }, "requirements": { "sk": [ { "kind": "weights", "text": "≈261,3 GB" } ], "en": [ { "kind": "weights", "text": "≈261.3 GB" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Jedna z vizuálne najkrajších scén benchmarku, ale podľa testu je hra nehrateľná a obsahuje množstvo chýb.", "en": "One of the benchmark's strongest-looking scenes, but testing found the game unplayable and full of bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/minimax-m3.png", "demo": "https://dubovskylabs.com/projects/baa/minimax-m3/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "MiniMaxAI/MiniMax-M3@f0e1c1e04d40177e4673a22097036854f536e9c0", "checkedAt": "2026-08-07", "sources": [ { "id": "minimax-m3-model", "kind": "official checkpoint", "url": "https://huggingface.co/MiniMaxAI/MiniMax-M3/tree/f0e1c1e04d40177e4673a22097036854f536e9c0", "revision": "f0e1c1e04d40177e4673a22097036854f536e9c0", "accessedAt": "2026-08-07" }, { "id": "minimax-m3-mxfp4", "kind": "vendor 4-bit conversion", "url": "https://huggingface.co/amd/MiniMax-M3-MXFP4/tree/b83d14e3d64bf373a207f3c2a7e9f0b0f1e7fc3a", "revision": "b83d14e3d64bf373a207f3c2a7e9f0b0f1e7fc3a", "accessedAt": "2026-08-07" }, { "id": "llama-cpp-q4-k-m", "kind": "runtime documentation", "url": "https://github.com/ggml-org/llama.cpp/blob/15586e2d7165570fb3aa7c26e0d442e289ef69de/tools/quantize/README.md", "revision": "15586e2d7165570fb3aa7c26e0d442e289ef69de", "accessedAt": "2026-08-07" } ] } }, { "rank": 18, "model": "Claude Sonnet 4.6", "publisher": "Anthropic", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Niekoľko chýb", "note": "Generovanie trvalo mimoriadne dlho, vyše hodiny." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Several issues", "note": "Generation took unusually long, over an hour." } }, "description": { "sk": "Vizuálne konzistentná hra s reálnym pohybom psa, stádom aj líškou; tvorba trvala dlhšie (~1,5 h).", "en": "Visually consistent game with real dog movement, a flock, and a fox; took longer to produce (~1.5 h)." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/claude-sonet.png", "demo": "https://dubovskylabs.com/projects/baa/claude-sonnet-4.6/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 19, "model": "GLM 5.2", "publisher": "Zhipu AI", "modelId": "glm-5-2", "runtime": { "sk": "Otvorené váhy (OpenRouter)", "en": "Open-weight (OpenRouter)" }, "parameters": { "sk": "744B", "en": "744B" }, "requirements": { "sk": [ { "kind": "weights", "text": "464,8 GB" }, { "kind": "context", "text": "≈9,52 GB / 100K" } ], "en": [ { "kind": "weights", "text": "464.8 GB" }, { "kind": "context", "text": "≈9.52 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Nehrateľný výstup; grafika je menej výrazná než pri GPT-5.5, ale kamera a ovládanie pôsobia o niečo stabilnejšie.", "en": "Unplayable output; visuals are weaker than GPT-5.5, but camera and controls feel somewhat more stable." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/glm-5-2.png", "demo": "https://dubovskylabs.com/projects/baa/glm-5.2/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "zai-org/GLM-5.2@b4734de4facf877f85769a911abafc5283eab3d9", "checkedAt": "2026-08-07", "sources": [ { "id": "glm-5-2-model", "kind": "official checkpoint", "url": "https://huggingface.co/zai-org/GLM-5.2/tree/b4734de4facf877f85769a911abafc5283eab3d9", "revision": "b4734de4facf877f85769a911abafc5283eab3d9", "accessedAt": "2026-08-07" }, { "id": "glm-5-2-nvfp4", "kind": "vendor 4-bit conversion", "url": "https://huggingface.co/nvidia/GLM-5.2-NVFP4/tree/aec724e8c7b8ee9db3b48c01c320f63f9cdaf8aa", "revision": "aec724e8c7b8ee9db3b48c01c320f63f9cdaf8aa", "accessedAt": "2026-08-07" }, { "id": "vllm-cache-layout", "kind": "runtime source", "url": "https://github.com/vllm-project/vllm/tree/946452961265514622b95ea170839b214f39c2a0", "revision": "946452961265514622b95ea170839b214f39c2a0", "accessedAt": "2026-08-07" } ] } }, { "rank": 20, "model": "DeepSeek V4 Flash", "publisher": "DeepSeek", "modelId": "deepseek-v4-flash", "runtime": { "sk": "Otvorené váhy (OpenRouter)", "en": "Open-weight (OpenRouter)" }, "parameters": { "sk": "284B", "en": "284B" }, "requirements": { "sk": [ { "kind": "weights", "text": "159,6 GB" }, { "kind": "context", "text": "≈0,71 GB / 100K" } ], "en": [ { "kind": "weights", "text": "159.6 GB" }, { "kind": "context", "text": "≈0.71 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Pekná, čistá arkádová scéna so stádom a domčekom, no prakticky nehrateľný výstup.", "en": "Clean, tidy arcade scene with a flock and a house, but a practically unplayable output." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/deepseek-v4-flash.png", "demo": "https://dubovskylabs.com/projects/baa/deepseek-v4-flash/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "deepseek-ai/DeepSeek-V4-Flash@60d8d70770c6776ff598c94bb586a859a38244f1", "checkedAt": "2026-08-07", "sources": [ { "id": "deepseek-v4-flash-model", "kind": "official checkpoint and inference code", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash/tree/60d8d70770c6776ff598c94bb586a859a38244f1", "revision": "60d8d70770c6776ff598c94bb586a859a38244f1", "accessedAt": "2026-08-07" } ] } }, { "rank": 21, "model": "Qwen 3.7 Max", "publisher": "Alibaba", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Veľmi veľa chýb", "note": "Vyžadoval 1 opravný prompt." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Very many bugs", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Kompletná 3D scéna sa spustí, ale výsledok je nehrateľný a obsahuje veľmi veľa chýb.", "en": "A complete 3D scene starts, but the result is unplayable and contains a large number of bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen-3-7-max.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.7-max/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": null }, { "rank": 22, "model": "DeepSeek V4 Pro", "publisher": "DeepSeek", "modelId": "deepseek-v4-pro", "runtime": { "sk": "Otvorené váhy (OpenRouter)", "en": "Open-weight (OpenRouter)" }, "parameters": { "sk": "1,6T", "en": "1.6T" }, "requirements": { "sk": [ { "kind": "weights", "text": "864,7 GB" }, { "kind": "context", "text": "≈1,01 GB / 100K" } ], "en": [ { "kind": "weights", "text": "864.7 GB" }, { "kind": "context", "text": "≈1.01 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval 1 opravný prompt." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Nehrateľný výsledok s opačne otočenou kamerou, množstvom chýb a jedným opravným promptom.", "en": "Unplayable result with reversed camera orientation, many bugs, and one corrective prompt." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/deepseek-v4-pro.png", "demo": "https://dubovskylabs.com/projects/baa/deepseek-v4-pro/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "deepseek-ai/DeepSeek-V4-Pro@b5968e9190ef611bbf34a7229255be88a0e937c1", "checkedAt": "2026-08-07", "sources": [ { "id": "deepseek-v4-pro-model", "kind": "official checkpoint and inference code", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/tree/b5968e9190ef611bbf34a7229255be88a0e937c1", "revision": "b5968e9190ef611bbf34a7229255be88a0e937c1", "accessedAt": "2026-08-07" } ] } }, { "rank": 23, "model": "Big Pickle", "publisher": "Unknown", "modelId": "big-pickle-unresolved", "runtime": { "sk": "Neznáme", "en": "Unknown" }, "parameters": { "sk": "—", "en": "—" }, "requirements": { "sk": [], "en": [] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Prostredie sa vykreslí, no model psa je rozbitý (čierny blok) a hra sa reálne nerozbehne.", "en": "The environment renders, but the dog model is broken (a black block) and the game never really starts." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/big-pickle.png", "demo": "https://dubovskylabs.com/projects/baa/big-pickle/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": null, "checkedAt": "2026-08-07", "sources": [] } }, { "rank": 24, "model": "Kimi 2.7 Code", "publisher": "Moonshot AI", "modelId": "kimi-k2-7-code", "runtime": { "sk": "Otvorené váhy (OpenRouter)", "en": "Open-weight (OpenRouter)" }, "parameters": { "sk": "1T", "en": "1T" }, "requirements": { "sk": [ { "kind": "weights", "text": "595,2 GB" } ], "en": [ { "kind": "weights", "text": "595.2 GB" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval 3 opravné prompty." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "Required 3 corrective prompts." } }, "description": { "sk": "Nehrateľný výsledok; pes aj kamera sú zle otočené a výstup potreboval tri opravné prompty.", "en": "Unplayable result; dog and camera orientation are wrong and the output needed three corrective prompts." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/kimi-2-7-code.png", "demo": "https://dubovskylabs.com/projects/baa/kimi-2.7-code/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "moonshotai/Kimi-K2.7-Code@74797c9c62378b951a1f6fcf5c4631024e9b8bef", "checkedAt": "2026-08-07", "sources": [ { "id": "kimi-k2-7-code-model", "kind": "official checkpoint", "url": "https://huggingface.co/moonshotai/Kimi-K2.7-Code/tree/74797c9c62378b951a1f6fcf5c4631024e9b8bef", "revision": "74797c9c62378b951a1f6fcf5c4631024e9b8bef", "accessedAt": "2026-08-07" } ] } }, { "rank": 25, "model": "Qwen 3.6 27B", "publisher": "Alibaba", "modelId": "qwen3-6-27b", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "27B", "en": "27B" }, "requirements": { "sk": [ { "kind": "weights", "text": "19,1 GB" }, { "kind": "context", "text": "≈6,71 GB / 100K" } ], "en": [ { "kind": "weights", "text": "19.1 GB" }, { "kind": "context", "text": "≈6.71 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Čiastočne splnené zadanie, jednoduchšia grafika a nehrateľný výsledok s množstvom chýb.", "en": "Partially matched assignment, simpler visuals, and an unplayable result with many bugs." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen3-6-27b.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.6-27b/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "Qwen/Qwen3.6-27B@6a9e13bd6fc8f0983b9b99948120bc37f49c13e9", "checkedAt": "2026-08-07", "sources": [ { "id": "qwen3-6-27b-config", "kind": "official config", "url": "https://huggingface.co/Qwen/Qwen3.6-27B/blob/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9/config.json", "revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9", "accessedAt": "2026-08-07" }, { "id": "qwen3-6-27b-q4", "kind": "runtime-project 4-bit checkpoint", "url": "https://huggingface.co/ggml-org/Qwen3.6-27B-GGUF/blob/8a7ee08e8b9bfb857107ecc25a5599d2f38b76f8/Qwen3.6-27B-Q4_K_M.gguf", "revision": "8a7ee08e8b9bfb857107ecc25a5599d2f38b76f8", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 26, "model": "Qwen 3.5 122B A10B", "publisher": "Alibaba", "modelId": "qwen3-5-122b-a10b", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "122B", "en": "122B" }, "requirements": { "sk": [ { "kind": "weights", "text": "78,9 GB" }, { "kind": "context", "text": "≈2,61 GB / 100K" } ], "en": [ { "kind": "weights", "text": "78.9 GB" }, { "kind": "context", "text": "≈2.61 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Veľmi veľa chýb", "note": "Ovce aj pes lietajú." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Very many bugs", "note": "Both the sheep and dog fly around." } }, "description": { "sk": "Herný svet sa vykreslí, no pes aj ovce lietajú a výsledok je nehrateľný.", "en": "The game world renders, but both the dog and sheep fly around and the result is unplayable." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen3-5-122b-a10b.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.5-122b-a10b/ovciak-game.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "Qwen/Qwen3.5-122B-A10B-GPTQ-Int4@30cd92cba9707a9aba09d1e490ed4b66b78e9606", "checkedAt": "2026-08-07", "sources": [ { "id": "qwen3-5-122b-config", "kind": "official config", "url": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B/blob/dc4d348443bc740c68e2d77492492c11606384d5/config.json", "revision": "dc4d348443bc740c68e2d77492492c11606384d5", "accessedAt": "2026-08-07" }, { "id": "qwen3-5-122b-int4", "kind": "official 4-bit checkpoint", "url": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B-GPTQ-Int4/tree/30cd92cba9707a9aba09d1e490ed4b66b78e9606", "revision": "30cd92cba9707a9aba09d1e490ed4b66b78e9606", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 27, "model": "GLM 4.7 Flash", "publisher": "Zhipu AI", "modelId": "glm-4-7-flash", "runtime": { "sk": "Otvorené váhy (OpenRouter)", "en": "Open-weight (OpenRouter)" }, "parameters": { "sk": "30B", "en": "30B" }, "requirements": { "sk": [ { "kind": "weights", "text": "≈19,1 GB" }, { "kind": "context", "text": "≈5,41 GB / 100K" } ], "en": [ { "kind": "weights", "text": "≈19.1 GB" }, { "kind": "context", "text": "≈5.41 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval 1 opravný prompt." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Prázdny tmavý svet s drobným rozbitým modelom psa a odsunutou kamerou; nehrateľné.", "en": "Empty dark world with a tiny broken dog model and a distant camera; unplayable." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/glm4-7-flash.png", "demo": "https://dubovskylabs.com/projects/baa/glm-4.7-flash/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "zai-org/GLM-4.7-Flash@7dd20894a642a0aa287e9827cb1a1f7f91386b67", "checkedAt": "2026-08-07", "sources": [ { "id": "glm-4-7-flash-model", "kind": "official checkpoint", "url": "https://huggingface.co/zai-org/GLM-4.7-Flash/tree/7dd20894a642a0aa287e9827cb1a1f7f91386b67", "revision": "7dd20894a642a0aa287e9827cb1a1f7f91386b67", "accessedAt": "2026-08-07" }, { "id": "vllm-cache-layout", "kind": "runtime source", "url": "https://github.com/vllm-project/vllm/tree/946452961265514622b95ea170839b214f39c2a0", "revision": "946452961265514622b95ea170839b214f39c2a0", "accessedAt": "2026-08-07" }, { "id": "llama-cpp-q4-k-m", "kind": "runtime documentation", "url": "https://github.com/ggml-org/llama.cpp/blob/15586e2d7165570fb3aa7c26e0d442e289ef69de/tools/quantize/README.md", "revision": "15586e2d7165570fb3aa7c26e0d442e289ef69de", "accessedAt": "2026-08-07" } ] } }, { "rank": 28, "model": "Qwen 3.6 35B A3B", "publisher": "Alibaba", "modelId": "qwen3-6-35b-a3b", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "35B", "en": "35B" }, "requirements": { "sk": [ { "kind": "weights", "text": "≈21,4 GB" }, { "kind": "context", "text": "≈2,11 GB / 100K" } ], "en": [ { "kind": "weights", "text": "≈21.4 GB" }, { "kind": "context", "text": "≈2.11 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval ešte 1 opravný prompt." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Nehrateľný výsledok s množstvom chýb; na spustenie vyžadoval ešte jeden opravný prompt.", "en": "Unplayable result with many bugs; required one corrective prompt to run." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen3-6-35b-a3b.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.6-35b-a3b/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "Qwen/Qwen3.6-35B-A3B@995ad96eacd98c81ed38be0c5b274b04031597b0", "checkedAt": "2026-08-07", "sources": [ { "id": "qwen3-6-35b-config", "kind": "official config", "url": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B/blob/995ad96eacd98c81ed38be0c5b274b04031597b0/config.json", "revision": "995ad96eacd98c81ed38be0c5b274b04031597b0", "accessedAt": "2026-08-07" }, { "id": "llama-cpp-q4-k-m", "kind": "runtime documentation", "url": "https://github.com/ggml-org/llama.cpp/blob/15586e2d7165570fb3aa7c26e0d442e289ef69de/tools/quantize/README.md", "revision": "15586e2d7165570fb3aa7c26e0d442e289ef69de", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 29, "model": "Gemma 4 31B", "publisher": "Alphabet", "modelId": "gemma-4-31b-it", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "30,7B", "en": "30.7B" }, "requirements": { "sk": [ { "kind": "weights", "text": "17,7 GB" }, { "kind": "context", "text": "≈9,03 GB / 100K" } ], "en": [ { "kind": "weights", "text": "17.7 GB" }, { "kind": "context", "text": "≈9.03 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Zadanie splnil len veľmi čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval ešte 1 opravný prompt." }, "en": { "assignment": "Matched the assignment only very partially", "playability": "Not playable", "bugs": "Many bugs", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Jednoduchý výstup so slabšou grafikou; zadanie splnil len veľmi čiastočne a vyžadoval opravný prompt.", "en": "Simple output with weaker visuals; matched the assignment only very partially and required a corrective prompt." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gamma4-31b.png", "demo": "https://dubovskylabs.com/projects/baa/gemma-4-31b/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "google/gemma-4-31B-it-qat-q4_0-gguf@59dde24573e7e61570dba08b18a2e1fe246955ed", "checkedAt": "2026-08-07", "sources": [ { "id": "gemma-4-31b-config", "kind": "official config", "url": "https://huggingface.co/google/gemma-4-31B-it/blob/842da3794eaa0b77d5f08bae87a17459d91ff475/config.json", "revision": "842da3794eaa0b77d5f08bae87a17459d91ff475", "accessedAt": "2026-08-07" }, { "id": "gemma-4-31b-q4", "kind": "official 4-bit checkpoint", "url": "https://huggingface.co/google/gemma-4-31B-it-qat-q4_0-gguf/tree/59dde24573e7e61570dba08b18a2e1fe246955ed", "revision": "59dde24573e7e61570dba08b18a2e1fe246955ed", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 30, "model": "Laguna S 2.1", "publisher": "Poolside", "modelId": "laguna-s-2-1", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "118B", "en": "118B" }, "requirements": { "sk": [ { "kind": "weights", "text": "71,9 GB" }, { "kind": "context", "text": "≈4,99 GB / 100K" } ], "en": [ { "kind": "weights", "text": "71.9 GB" }, { "kind": "context", "text": "≈4.99 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Nesplnil zadanie", "playability": "Nehrateľný", "bugs": "Veľmi veľa vážnych chýb; padá", "note": "Vyžadoval 5 opravných promptov; prvé generovanie trvalo 1 h 31 min." }, "en": { "assignment": "Did not match the assignment", "playability": "Unplayable", "bugs": "Very many serious bugs; crashes", "note": "Required 5 corrective prompts; the first generation took 1 h 31 min." } }, "description": { "sk": "Ambiciózny šesťlevelový výstup so štyrmi plemenami a viacerými akciami, ale zadanie nesplnil a je nehrateľný: padá, pes sa pohybuje iba jedným smerom a postavy sú rozbité.", "en": "An ambitious six-level output with four breeds and several actions, but it did not match the assignment and is unplayable: it crashes, the dog moves in only one direction, and the characters are broken." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/laguna-s-2-1-118b.png", "demo": "https://dubovskylabs.com/projects/baa/laguna-s2.1-118b/index.html", "provenance": { "generatedAt": "2026-08-08", "testedAt": "2026-08-08", "promptVersion": "baa-original-sk", "runType": "multi-prompt" }, "modelSource": { "modelRevision": "poolside/Laguna-S-2.1-NVFP4-mlx@e6a961c3bbebfffd8fa5b42f243e375504f41edd", "checkedAt": "2026-08-08", "sources": [ { "id": "laguna-s-2-1-base", "kind": "official checkpoint and architecture", "url": "https://huggingface.co/poolside/Laguna-S-2.1/tree/00af5a51782109b587a3b3bbf11875e566036fa7", "revision": "00af5a51782109b587a3b3bbf11875e566036fa7", "accessedAt": "2026-08-08" }, { "id": "laguna-s-2-1-nvfp4-mlx", "kind": "official 4-bit MLX checkpoint", "url": "https://huggingface.co/poolside/Laguna-S-2.1-NVFP4-mlx/tree/e6a961c3bbebfffd8fa5b42f243e375504f41edd", "revision": "e6a961c3bbebfffd8fa5b42f243e375504f41edd", "accessedAt": "2026-08-08" } ] } }, { "rank": 31, "model": "GPT-5.6 Luna", "publisher": "OpenAI", "modelId": null, "runtime": { "sk": "Uzavretý", "en": "Closed" }, "parameters": null, "requirements": null, "evaluation": { "sk": { "assignment": "Nesplnil zadanie", "playability": "Hrateľný", "bugs": "Pár chýb", "note": "Zaznamenaný čas: 13 min 26 s." }, "en": { "assignment": "Did not match the assignment", "playability": "Playable", "bugs": "A few bugs", "note": "Recorded time: 13 min 26 sec." } }, "description": { "sk": "Vizuálne uhladená a hrateľná pseudo-3D arkáda s piatimi levelmi a tromi plemenami, ktorá však nesplnila požiadavku na skutočnú 3D hru.", "en": "A visually polished and playable pseudo-3D arcade game with five levels and three breeds, but it did not meet the requirement for a true 3D game." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/gpt-5-6-luna.png", "demo": "https://dubovskylabs.com/benchmarks/baa/apps/gpt-5-6-luna/index.html", "provenance": { "generatedAt": "2026-08-08", "testedAt": "2026-08-08", "promptVersion": "baa-original-sk", "runType": "one-shot" }, "modelSource": null }, { "rank": 32, "model": "MiMo v5", "publisher": "Xiaomi", "modelId": "mimo-v5-unresolved", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "—", "en": "—" }, "requirements": { "sk": [], "en": [] }, "evaluation": { "sk": { "assignment": "Splnil zadanie čiastočne", "playability": "Nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval 1 opravný prompt." }, "en": { "assignment": "Partially matched the assignment", "playability": "Not playable", "bugs": "Many bugs", "note": "Required 1 corrective prompt." } }, "description": { "sk": "Scéna existuje, no svet je extrémne tmavý a takmer nič v ňom nevidno; nehrateľné.", "en": "A scene exists, but the world is extremely dark and almost nothing is visible; unplayable." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/mimo-v5.png", "demo": "https://dubovskylabs.com/projects/baa/mimo-v5/sheepdog.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": null, "checkedAt": "2026-08-07", "sources": [] } }, { "rank": 33, "model": "Mistral Small 4", "publisher": "Mistral AI", "modelId": "mistral-small-4-119b", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "119B", "en": "119B" }, "requirements": { "sk": [ { "kind": "weights", "text": "70,8 GB" }, { "kind": "context", "text": "≈2,3 GB / 100K" } ], "en": [ { "kind": "weights", "text": "70.8 GB" }, { "kind": "context", "text": "≈2.3 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Nesplnil zadanie", "playability": "Absolútne nehrateľný", "bugs": "Veľmi veľa vážnych chýb", "note": "Vyžadoval 3 opravné prompty." }, "en": { "assignment": "Did not match the assignment", "playability": "Completely unplayable", "bugs": "Very many serious bugs", "note": "Required 3 corrective prompts." } }, "description": { "sk": "Tmavá a zle čitateľná scéna s veľkými objektmi pred kamerou; výsledok je absolútne nehrateľný.", "en": "Dark, hard-to-read scene with large objects blocking the camera; the result is completely unplayable." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/mistral-small-4.png", "demo": "https://dubovskylabs.com/projects/baa/mistral-small-4/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "mistralai/Mistral-Small-4-119B-2603-NVFP4@b1a9048590131d38491bd23a7c9f6ed0962f0358", "checkedAt": "2026-08-07", "sources": [ { "id": "mistral-small-4-nvfp4", "kind": "official 4-bit checkpoint and deployment recipe", "url": "https://huggingface.co/mistralai/Mistral-Small-4-119B-2603-NVFP4/tree/b1a9048590131d38491bd23a7c9f6ed0962f0358", "revision": "b1a9048590131d38491bd23a7c9f6ed0962f0358", "accessedAt": "2026-08-07" } ] } }, { "rank": 34, "model": "Ornith 1.0 35B", "publisher": "Ornith AI", "modelId": "ornith-1-0-35b", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "35B", "en": "35B" }, "requirements": { "sk": [ { "kind": "weights", "text": "21,2 GB" }, { "kind": "context", "text": "≈2,11 GB / 100K" } ], "en": [ { "kind": "weights", "text": "21.2 GB" }, { "kind": "context", "text": "≈2.11 GB / 100K" } ] }, "evaluation": { "sk": { "assignment": "Nesplnil zadanie", "playability": "Absolútne nehrateľný", "bugs": "Množstvo chýb", "note": "Vyžadoval 2 opravné prompty." }, "en": { "assignment": "Partially matched the assignment", "playability": "Completely unplayable", "bugs": "Many bugs", "note": "Required 2 corrective prompts." } }, "description": { "sk": "Kamera mieri mimo hracej plochy a herný svet takmer nevidno; absolútne nehrateľné aj po dvoch opravných promptoch.", "en": "The camera points away from the play area and the game world is barely visible; completely unplayable even after two corrective prompts." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/ornith-1-0-35b.png", "demo": "https://dubovskylabs.com/projects/baa/ornith-1.0-35b/index.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "ornith-ai/Ornith-1.0-35B-GGUF@383064f72a1ef3087b779f268d3ca117eb989aac", "checkedAt": "2026-08-07", "sources": [ { "id": "ornith-35b-model", "kind": "publisher checkpoint", "url": "https://huggingface.co/ornith-ai/Ornith-1.0-35B/tree/5df2ed3f675c7beaa490328cc70bb573b65fb660", "revision": "5df2ed3f675c7beaa490328cc70bb573b65fb660", "accessedAt": "2026-08-07" }, { "id": "ornith-35b-q4", "kind": "publisher 4-bit checkpoint", "url": "https://huggingface.co/ornith-ai/Ornith-1.0-35B-GGUF/tree/383064f72a1ef3087b779f268d3ca117eb989aac", "revision": "383064f72a1ef3087b779f268d3ca117eb989aac", "accessedAt": "2026-08-07" }, { "id": "transformers-cache", "kind": "runtime source", "url": "https://github.com/huggingface/transformers/tree/42f189ded85d18d00b51161d694cafd325e32b91", "revision": "42f189ded85d18d00b51161d694cafd325e32b91", "accessedAt": "2026-08-07" } ] } }, { "rank": 35, "model": "Qwen 3.5 A3B", "publisher": "Alibaba", "modelId": "qwen3-5-a3b-unresolved", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "—", "en": "—" }, "requirements": { "sk": [], "en": [] }, "evaluation": { "sk": { "assignment": "Nesplnil zadanie", "playability": "Absolútne nehrateľný", "bugs": "Množstvo chýb", "note": "" }, "en": { "assignment": "Partially matched the assignment", "playability": "Completely unplayable", "bugs": "Many bugs", "note": "" } }, "description": { "sk": "Najhorší z výstupov, ktoré niečo vygenerovali: zobrazí sa len UI na prázdnom pozadí, herný svet nevidno.", "en": "Worst of the outputs that produced anything: only the UI shows on an empty background, the game world is not visible." }, "screenshot": "https://dubovskylabs.com/assets/img/benchmarks/baa/qwen3-5-a3b.png", "demo": "https://dubovskylabs.com/projects/baa/qwen-3.5-a3b/sheep-herder.html", "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": null, "checkedAt": "2026-08-07", "sources": [ { "id": "qwen3-5-35b-model", "kind": "official checkpoint", "url": "https://huggingface.co/Qwen/Qwen3.5-35B-A3B", "revision": "main checked 2026-08-07", "accessedAt": "2026-08-07" }, { "id": "qwen3-30b-model", "kind": "official checkpoint", "url": "https://huggingface.co/Qwen/Qwen3-30B-A3B", "revision": "ad44e77", "accessedAt": "2026-08-07" } ] } }, { "rank": 36, "model": "Nemotron 3 Ultra", "publisher": "NVIDIA", "modelId": "nemotron-3-ultra", "runtime": { "sk": "Otvorené váhy", "en": "Open-weight" }, "parameters": { "sk": "550B", "en": "550B" }, "requirements": { "sk": [ { "kind": "weights", "text": "352,3 GB" } ], "en": [ { "kind": "weights", "text": "352.3 GB" } ] }, "evaluation": { "sk": { "assignment": "Nesplnil zadanie", "playability": "Nič nevygeneroval", "bugs": "—", "note": "" }, "en": { "assignment": "Did not match the assignment", "playability": "Produced nothing", "bugs": "—", "note": "" } }, "description": { "sk": "Nemotron 3 Ultra nevygeneroval žiadny výstup", "en": "Nemotron 3 Ultra produced no output" }, "screenshot": null, "demo": null, "provenance": { "generatedAt": null, "testedAt": null, "promptVersion": null, "runType": null }, "modelSource": { "modelRevision": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4@183968f87ae4cedce3039313cac1fd43d112c578", "checkedAt": "2026-08-07", "sources": [ { "id": "nemotron-3-ultra-nvfp4", "kind": "official 4-bit checkpoint", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4/tree/183968f87ae4cedce3039313cac1fd43d112c578", "revision": "183968f87ae4cedce3039313cac1fd43d112c578", "accessedAt": "2026-08-07" } ] } } ] }