| {: , : , : , : Boxty v1.0\tech vocabulary\WER 1.35%\Jachnun v1.0\code-switching\WER 5.85%\teudat zehut\arnona\makolet\GitHub Gist\Kubernetes\Sabich v2.2\Code-switching WER 4.91%\Tech WER 2.76%\accuracy preserved!\, : null, : [, , , , , , , , , , , , , , ], : [, , , ], : [, , , , , , ], : } |
| {: , : , : , : Backup 3 of 7 complete\DO NOT UNPLUG\NAS\, : null, : [, , ], : [, ], : [, , , , ], : } |
| {: , : , : , : $ claude plugin install csv-tools\, : null, : [], : [, ], : [, , , ], : } |
| {: , : , : , : 18 Central Station 4 min\74 Malha Mall 11 min\, : null, : [, , , , , ], : [, ], : [, , , , ], : } |
| {: , : , : , : Daniel's MCP\" glowing green-to-cyan, second line \"server collection\" glowing orange-to-pink. A small pixel-art sloth face sits centred above the text. No other text, logos or lettering anywhere in the image.", "input_image": null, "expected_text": ["Daniel's MCPserver collectionvalidationtext_exacttext_legiblestructure_checkstwo lines in the stated orderline colours as asked (green-to-cyan, then orange-to-pink)pixel-art sloth face centred aboveno other text, including on the circuit boardsnotesModelled on the Claude-Code-Plugins README banner. The apostrophe counts toward text_exact. |
| idtgb-t2i-10modalitytext-to-imagestyleeditorial lineage diagrampromptA 16:9 model-lineage diagram for a blog post. Calm editorial style for a personal tech blog: warm cream background, generous white space, thin slate-grey lines, a muted palette of deep teal, amber and slate, clean modern sans-serif text with a serif for the heading. Flat, no 3D, no gradients. Serif heading at the top left: \. On the left, one box labelled \ with a smaller grey line under it: \. From that box two arrows run to the right, one up to an upper row and one down to a lower row. Upper row, three small pale boxes joined by plus signs, in this order: \, \, \, then an arrow to a teal badge reading \ with \ in small text under it. Lower row, in this order: \, \, \, then an arrow to an amber badge reading \ with \ under it. No other text, logos or watermarks anywhere.input_imagevalidationtext_exacttext_legiblestructure_okmeaning_preservedexpected_textOne base, two recipesWhisper Smallstock modelTech termsOwn voiceReplayBoxty v1.0210 minCode-switchingJachnun v1.0483 minstructure_checksone base box on the lefttwo arrows from it: one to the upper row, one to the lower rowupper row order: Tech terms + Own voice + Replay -> Boxty v1.0lower row order: Code-switching + Own voice + Replay -> Jachnun v1.0Boxty badge teal, Jachnun badge amber, minutes under the right badgenotesModelled on the lineage figure in the Jachnun/Sabich post on blog.danielrosehill.com. 'Own voice' and 'Replay' appear twice each; text_exact needs both copies. Minutes are illustrative. |
| idtgb-t2i-11modalitytext-to-imagestyleeditorial title cardpromptA 16:9 social share card for a blog post. Calm editorial style for a personal tech blog: warm cream background, generous white space, thin slate-grey lines, a muted palette of deep teal, amber and slate, clean modern sans-serif text with a serif for the heading. Flat, no 3D, no gradients. Large serif title at the top left: \. Under it, in smaller sans-serif: \ Below that, a white rounded card with a thin border holding two rows. Row one: a small grey label \ on the left, and in muted red: \. Row two: a small teal label \ on the left, and in bold teal: \. No other text, logos or watermarks anywhere.input_imagevalidationtext_exacttext_legiblestructure_okmeaning_preservedexpected_textSabich v2.2One model for tech jargon and Hebrew words.STOCKI need to renew my to the zehutFINE-TUNEDI need to renew my teudat zehutstructure_checkstitle, subtitle, then the card, top to bottomSTOCK row above FINE-TUNED rowstock line in red, fine-tuned line bold tealthe deliberately wrong 'to the zehut' kept as written, not correctednotesModelled on the Hebrish share card (STOCK vs HEBRISH rows) on blog.danielrosehill.com. The stock line is a deliberate mis-transcription: a model that 'fixes' it fails text_exact. |
| idtgb-t2i-12modalitytext-to-imagestyleinformation tablepromptA 16:9 information table drawn as a clean graphic for a blog post. Calm editorial style for a personal tech blog: warm cream background, generous white space, thin slate-grey lines, a muted palette of deep teal, amber and slate, clean modern sans-serif text with a serif for the heading. Flat, no 3D, no gradients. Serif title above the table: \. Four columns with a deep-teal header row in white text, in this order: \, \, \, \. Three body rows with thin rules between them, in this order: row 1 \, \, \, \; row 2 \, \, \, \; row 3 \, \, \, \. Highlight the Sabich v2.2 row with a pale amber fill. Numbers right-aligned. Under the table, in small grey text: \ No other text, logos or watermarks anywhere.input_imagevalidationtext_exacttext_legiblestructure_okmeaning_preservedexpected_textHebrish models at a glanceModelTrained onTech WERCode-switching WERBoxty v1.0Tech vocabulary1.35%9.80%Jachnun v1.0Code-switching3.12%5.85%Sabich v2.2Both2.76%4.91%Lower is better.structure_checksexactly 4 columns in the stated orderexactly 3 body rows in the stated orderevery value in its own row and column (no swapped cells)teal header, amber Sabich rowno extra rows, columns or totalsnotesTables are a common blog ask. 1.35, 5.85, 2.76 and 4.91 are the real figures from tgb-t2i-03; 9.80 and 3.12 are illustrative. A cell in the wrong column fails structure_ok even if spelled right. |
| idtgb-t2i-13modalitytext-to-imagestylenumbered-steps infographicpromptA 16:9 infographic for a blog post showing a five-step process from left to right. Calm editorial style for a personal tech blog: warm cream background, generous white space, thin slate-grey lines, a muted palette of deep teal, amber and slate, clean modern sans-serif text with a serif for the heading. Flat, no 3D, no gradients. Serif title at the top: \. Five numbered circles in a row, numbered 1 to 5, joined by thin arrows pointing right. Above each circle a simple flat line icon; below each, a bold label and a one-line caption in grey. In order: 1 \ / \; 2 \ / \; 3 \ / \; 4 \ / \; 5 \ / \. No other text, logos or watermarks anywhere.input_imagevalidationtext_exacttext_legiblestructure_okmeaning_preservedexpected_textFrom voice note to blog post12345RecordPhone voice memoTranscribeFine-tuned ASRClean upParagraphs and punctuationDraftLLM with my style guidePublishAstro site on Vercelstructure_checksfive steps, numbered 1-5 left to rightarrows between consecutive steps, all pointing righteach caption under its own labelan icon above each circle that fits the stepnotesNumbered-step infographics are the other common informational ask. Icons with embedded text (e.g. a fake app name) count as invented text. |
| idtgb-t2i-14modalitytext-to-imagestylebar chartpromptA 16:9 horizontal bar chart for a blog post. Calm editorial style for a personal tech blog: warm cream background, generous white space, thin slate-grey lines, a muted palette of deep teal, amber and slate, clean modern sans-serif text with a serif for the heading. Flat, no 3D, no gradients. Serif title at the top left: \. Five horizontal bars, top to bottom in this order, each with its label on the left and its value at the end of the bar: \ 8.4%, \ 6.1%, \ 4.9%, \ 2.76%, \ 1.35%. Bar lengths proportional to the values on a shared axis starting at zero. Stock models in slate grey, the two fine-tunes in teal. Light vertical gridlines; x-axis label under the chart: \. No other text, logos or watermarks anywhere.input_imagevalidationtext_exacttext_legiblestructure_okmeaning_preservedexpected_textWord error rate on tech vocabularyWhisper Small (stock)8.4%Whisper Medium (stock)6.1%Parakeet TDT (stock)4.9%Sabich v2.22.76%Boxty v1.01.35%WER (%)structure_checksfive bars in the stated order, longest at the topbar lengths roughly proportional (8.4 about six times 1.35)each value at the end of its own barstock bars grey, fine-tune bars tealaxis starts at zero; any tick numbers must be correctnotesCharts from an image model are a stretch, but the blog uses them (Knesset forecast dot plot). Tick numbers are allowed if correct; wrong ticks fail text_exact. Stock figures are illustrative. |
| idtgb-i2i-01modalityimage-to-imagestylekeep handwriting, add colourpromptTurn this whiteboard photo into a clean, colourful diagram. Remove the whiteboard frame, glare and background, and correct the perspective to a straight-on view on a white background. Keep my handwriting style but make it neater and easy to read. Keep every label, box and arrow exactly where it is and pointing the same way. Fix obvious misspellings of technical names (\, \, \) but do not add, remove or reword any label. Add colour: give each group of related boxes its own soft colour and make the arrows a contrasting colour.input_imageinputs/whiteboard-mcp-01.jpgexpected_textWANTelegramCF / TailscaleOpenClawDockerVMMETA MCPMETA MCP 2Desktop ClientNo proxy possibleLAN useLAN MCPsNASHASBCsProxmoxSSE Remote MCPsReplicateGooglevalidationtext_exacttext_legiblestructure_okmeaning_preservedstructure_checksDesktop Client -> META MCP 2Desktop Client -> VMWAN -> OpenClaw (labelled CF / Tailscale)META MCP -> LAN MCPsMETA MCP -> SSE Remote MCPsLAN MCPs -> NAS, HA, SBCs, ProxmoxSSE Remote MCPs -> Replicate, Googleupward arrow labelled LAN usehandwriting character keptno labels added or dropped |
| idtgb-i2i-03modalityimage-to-imagestylechalkboardpromptTurn this whiteboard photo into a diagram drawn on a classic green chalkboard. Remove the physical whiteboard, its frame, the pen tray, glare and the wall. Correct the perspective so the result is a straight-on view. Use white chalk for the text and coloured chalk (yellow, pink, light blue) for the boxes and arrows, with subtle chalk dust and smudges. The lettering should be neat chalk handwriting. Preserve every word, box and arrow, and do not add, remove or reword any text: copy each word as written, correcting only what the handwriting makes ambiguous. Keep every arrow pointing the same way between the same items.input_imageinputs/whiteboards-08.jpgexpected_textFew Shot Learning - 3Zoom Meeting Prep SOPIs my hair a mess?YNCOMB HAIRCHILL OUT!validationtext_exacttext_legiblestructure_okmeaning_preservedstructure_checksquestion branches into Y on the left and N on the rightY -> COMB HAIR, N -> CHILL OUT!COMB HAIR and CHILL OUT! are boxedtitle underlined at top leftgreen chalkboard looknotesInput: https://huggingface.co/datasets/danielrosehill/Whiteboards sample 08 (flowchart). expected_text from its human transcription; case-insensitive (the board mixes caps). |
| idtgb-i2i-04modalityimage-to-imagestylesticky notespromptTurn this whiteboard photo into a colourful sticky-note board. Remove the physical whiteboard, its frame, the pen tray, glare and the wall. Correct the perspective so the result is a straight-on view. Put it on a light cork-board background. Render each block of text and each box as its own sticky note (yellow, pink, blue, green) with a slight shadow and tilt, in a casual handwritten font, and draw the arrows as marker lines between the notes. Keep the little bot and smiley faces as small doodles. Preserve every word, box and arrow, and do not add, remove or reword any text: copy each word as written, correcting only what the handwriting makes ambiguous. Keep every arrow pointing the same way between the same items.input_imageinputs/whiteboards-11.jpgexpected_textPERSONALISED AIBASIC IDEAUse-case:Human wants to create a task-specific AI agent& frontload contextTactic:Create AI agent to \ youtranscribe(could be voice → voice!)extract contentvector DBWhat's your fav food?", "PIZZA!", "CREATE MEMORY", "Daniel's fav food = PIZZAVECTOR DBvalidationtext_exacttext_legiblestructure_okmeaning_preservedstructure_checksbot face -> speech bubble What's your fav food?", "PIZZA! reply next to the smiley face", "bot -> CREATE MEMORY box, PIZZA! -> memory box", "memory box -> VECTOR DB cylinder", "tactic line keeps its arrow chain in order", "bot and smiley faces kept"], "notes": "Input: https://huggingface.co/datasets/danielrosehill/Whiteboards sample 11 (diagram). 16 strings; case-insensitive. A dense board: tests whether splitting text across notes drops or rewords any of it."} |
| {"id": "tgb-i2i-05", "modality": "image-to-image", "style": "comic book", "prompt": "Turn this whiteboard photo into a comic-book page. Remove the physical whiteboard, its frame, the pen tray, glare and the wall. Correct the perspective so the result is a straight-on view. Bold black ink outlines, halftone dot shading and a bright red, blue, yellow and green palette. Draw the smiley face as a cheerful human character and the box-headed face as a friendly robot, and put each line of dialogue in a speech bubble from the right character. Add action lines for energy, but no sound-effect words or any other extra text. Preserve every word, box and arrow, and do not add, remove or reword any text: copy each word as written, correcting only what the handwriting makes ambiguous. Keep every arrow pointing the same way between the same items.", "input_image": "inputs/whiteboards-13.jpg", "expected_text": ["PERSONALISED AI", "WORKFLOW 2", "Hey bot! Let's build out my food & drink context store!Sure! Let me see what gaps we have...READvector DBHmm. nothing about beer preferences!What beer do you prefer?Super bitter IPAsCREATEUser's fav beer = bitter IPA"], "validation": ["text_exact", "text_legible", "structure_ok", "meaning_preserved"], "structure_checks": ["user lines come from the human, bot lines from the robot", "dialogue order unchanged top to bottom", "READ -> vector DB -> bot (Hmm...)", "CREATE arrow leads to the memory oval User's fav beer = bitter IPAno POW/ZAP or other invented wordsnotesInput: https://huggingface.co/datasets/danielrosehill/Whiteboards sample 13 (diagram). Case-insensitive. The comic style pulls towards sound-effect words, which count as invented text. |
| idtgb-i2i-06modalityimage-to-imagestyleneon signpromptTurn this whiteboard photo into a neon-sign installation. Remove the physical whiteboard, its frame, the pen tray, glare and the wall. Correct the perspective so the result is a straight-on view. Put it on a dark exposed-brick wall. Render every line, arrow and word as a glowing neon tube (pink, blue, white and yellow), with soft light bleeding onto the bricks; arrows are continuous bent tubes. Preserve every word, box and arrow, and do not add, remove or reword any text: copy each word as written, correcting only what the handwriting makes ambiguous. Keep every arrow pointing the same way between the same items.input_imageinputs/whiteboards-21.jpgexpected_textWhiteboardDev SpecAgent DevWhiteboard(s)Gemini APIGeminiCleaned Tech Diagrams(image 2 image)Narrative of Agent Model(image 2 txt)Agent 2:Convert these thingsPlan def w/ humanGenerate tasks, define subagents,checkpoints etcvalidationtext_exacttext_legiblestructure_okmeaning_preservedstructure_checkstitle chain Whiteboard -> Dev Spec -> Agent DevWhiteboard(s) -> Gemini APIGemini forks to Cleaned Tech Diagrams and Narrative of Agent ModelConvert these things -> Dev Spectwo sub-bullets under Agent 2 in ordernotesInput: https://huggingface.co/datasets/danielrosehill/Whiteboards sample 21 (diagram). Case-insensitive. Neon with this much text is the hardest style here for legibility. |
| idtgb-he-00modalitytext-to-imagetrackhebrewstyleflat colour, one labelpromptA single rounded box drawn in bright flat colours on a white background, like one element of a tech diagram. Inside the box, one Hebrew word in large clear type, written right to left exactly as given: \. Below the box, a smaller label in Hebrew: \. No other text.input_imageexpected_textפירגוןמילון מונחיםvalidationtext_exacttext_legiblestructure_okmeaning_preservedstructure_checksone boxHebrew word inside the box, label below itletters in the right order (right to left), not mirroredfinal letters (ן, ם) in their final formsno extra textnotesScreening prompt for the Hebrew track: one generation per model. Only models that pass text_exact here go on to tgb-he-01 and tgb-he-02. פירגון is carried over from Hebrew-Image-Generation-Eval (2025-12), where only Gemini 3 Pro, Nano Banana Pro and Wan 2.5 rendered it. |
| idtgb-he-01modalitytext-to-imagetrackhebrewstylebold flat colour, Hebrew onlypromptCreate a bright, colourful flat-colour explainer diagram in Hebrew, laid out for a right-to-left reader. All text is Hebrew, written right to left, spelled exactly as given. Title at the top: \. Four boxes in a single row, reading from right to left, connected by arrows pointing left: \ -> \ -> \ -> \. So \ is the rightmost box and \ the leftmost. Below the row, a separate box labelled \ with one arrow pointing up into \ only. Under \, a small badge reading \. A small, friendly cartoon robot mascot on the right edge points at \. At the bottom, a caption banner reading exactly: \. No other text.input_imageexpected_textאיך עובד זיהוי דיבורהקלטהתמלולתיקון טקסטטקסט סופימילון מונחיםשיעור שגיאות 2.76%מודל קטן, דיוק גבוהvalidationtext_exacttext_legiblestructure_okmeaning_preservedstructure_checksfour boxes in one row in the stated right-to-left orderthree left-pointing arrows between themמילון מונחים sits below the rowexactly one arrow from מילון מונחים, pointing up into תמלולbadge under טקסט סופיcharacter on the right edge pointing at הקלטהHebrew letters not mirrored or reversed; final forms (ך, ם, ן) correctthe percentage reads 2.76% (digits left to right, % in place)no extra textnotesHebrew-only mirror of tgb-t2i-02. 2.76% is Sabich v2.2's tech WER (see tgb-t2i-03). Common failures to log in text_errors: reversed letter order, mirrored glyphs, non-final letter at word end, Latin or pseudo-Hebrew filler, digits reversed by the bidi algorithm (6.72%)."} |
| {"id": "tgb-he-02", "modality": "text-to-image", "track": "hebrew", "style": "flat colour, mixed Hebrew and English", "prompt": "Create a tech diagram with clean bold lines and flat bright colours, mixing Hebrew and English text. Every string must be spelled exactly as given: Hebrew right to left, English left to right, and where they share a line each keeps its own direction. Title at the top: \"תמלול Hebrish\". Three boxes in a row, laid out right to left, with arrows pointing left: \"הקלטה\" (rightmost) -> \"Sabich v2.2\" -> \"Transcript\" (leftmost). Under \"Sabich v2.2\", a small subtitle: \"מודל מכוונן\". A small, friendly cartoon robot mascot on the right, next to \"הקלטה\", has a speech bubble saying exactly: \"יש לי meeting ב-Zoom מחר\". Under \"Transcript\", a sticky note reading: \"teudat zehut = תעודת זהות\". At the bottom, a caption banner: \"WER 4.91% ב-code-switching\". No other text.", "input_image": null, "expected_text": ["תמלול Hebrish", "הקלטה", "Sabich v2.2", "Transcript", "מודל מכוונן", "יש לי meeting ב-Zoom מחר", "teudat zehut = תעודת זהות", "WER 4.91% ב-code-switching"], "validation": ["text_exact", "text_legible", "structure_ok", "meaning_preserved"], "structure_checks": ["three boxes in one row in the stated right-to-left order", "two left-pointing arrows between them", "subtitle under Sabich v2.2", "speech bubble comes from the character beside הקלטה", "sticky note under Transcript", "in mixed strings each run keeps its own direction: English words not reversed, Hebrew words not reversed, word order in the sentence correct for a Hebrew reader", "Hebrew prefix ב- attached to Zoom and code-switching, not detached or moved", "no extra text"], "notes": "Code-switching subtask: Hebrew and English in the same string, as in Hebrish speech. 4.91% is Sabich v2.2's code-switching WER (see tgb-t2i-03). Grade word order inside mixed strings as a text error, not a structure error: 'מחר Zoom-ב meeting לי יש' is a fail. |
| |
