Open Weight Bench
← Overview

qwen/qwen3.6-27b@q8_0

qwen 27B · dense gguf / q8_0 ctx 256k released 2026-04-21 vision tool_use

coding

all models in this bench →
Error: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions' For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
Score
error
tok/s
0
Tokens
—
Wall
900.0 s
RAM
—
Worum geht es? Was wird getestet?
Task: From a ~200-word prompt the model must generate a fully functional Kanban board as a single-file HTML with drag & drop, localStorage persistence, edit/delete and a confetti animation — in a single chat without iteration. The prompt also includes a small `data-testid` contract so a Playwright test can drive the app remotely. Three signals feed into the score: (1) Static — a linter checks concrete constraints in the HTML (columns, Tailwind, localStorage call, no framework, no window.alert/prompt, …). (2) Functional — Playwright runs a small CRUD sequence: create a card, delete a card with confirmation, reload — does state persist? — and checks whether any JS console errors occur during the entire flow. Drag & drop and confetti are deliberately not tested functionally (too many implementation variants). (3) Qualitative — LLM-as-judge rates screenshot and code (visual + code quality + render↔code consistency). Score = mean over the available signals. Why models fail: reasoning models burn their tokens in thinking instead of writing. Sliding-window models (Gemma 4) lose the constraints at the start of the prompt. Small models (<3B) often fail to produce coherent HTML — or ignore the data-testid contract, which makes the functional tests fail in droves.
Prompt
System prompt
You are a careful front-end engineer.
Developer prompt
Create a fully functional Kanban board in a single HTML file using vanilla JavaScript (no frameworks like react).

Requirements:
- Columns: Backlog, In Progress, Review, Done.
- Cards must be:
  - draggable across columns,
  - editable in place,
  - persisted in localStorage (state survives reloads) - please use your own namespace,
  - deletable with a confirmation prompt.
- Each column provides an "Add card" action.
- Style with Tailwind via CDN.
- Add subtle CSS transitions and trigger a confetti animation when a card moves to "Done".
- Thoroughly comment the code.
- dont use window.alert or window.prompt to add/edit/delete cards
- if there are no cards yet, create some dummy cards
- modern and vibrant design

Stable test selectors (mandatory — these data-testid attributes are used by an automated functional test; do not omit, rename, or split them across multiple elements):
- Column containers: data-testid="column-backlog", data-testid="column-in-progress", data-testid="column-review", data-testid="column-done".
- Every "Add card" button (one per column): data-testid="add-card".
- Every card element: data-testid="card".
- Inside each card, the delete trigger: data-testid="delete-card".
- The confirm button of the delete-confirmation dialog/modal: data-testid="confirm-delete".
- The input/textarea where a new card title is typed: data-testid="card-input". Pressing Enter in this input MUST commit the new card.

As answer return the plain HTML of the working application (script and styles included)
▶ Open generated app in a new tab ↗
Artefacts: Generierte App Raw model response

vision

all models in this bench →
Score
0%
tok/s
0
Tokens
—
Wall
0.0 s
RAM
—
Worum geht es? Was wird getestet?
Task: Four OCR sub-tasks, one image each. (1)–(3) Handwritten meeting notes in three difficulty tiers (easy / medium / hard) — the model must transcribe the text. (4) An old book page in Fraktur typeface — same task. What is tested: OCR quality, recognising layout structure (columns, bullet points, dates), handling of illegible handwriting and historical letterforms (long-s, ligatures). Why models fail: text-only models have no vision capability (filtered out). Weak VLMs only recognise the clearest part. Some truncate output or get stuck in reasoning without producing a visible answer.
Prompt
System prompt
Du bist OCR-Spezialist für deutsche Handschrift.
Developer prompt
Auf dem Bild siehst du eine handschriftliche Meeting-Notiz mit klarer Struktur und gut lesbarer Schrift. Transkribiere den gesamten lesbaren Text. Behalte die Anordnung bei (Überschrift, Spalten, To-Dos). Bei unleserlichen Stellen schreibe '[unleserlich]'. Gib ausschließlich den puren OCR-Text im Markdown-Format zurück — keine Vorbemerkung, keine Erklärungen, kein Code-Fence.
Handschrift OCR – leicht
Recall 0%
Precision 0%
0/77 +0
—
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions' For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
Handschrift OCR – mittel
Recall 0%
Precision 0%
0/136 +0
—
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions' For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
Handschrift OCR – schwer
Recall 0%
Precision 0%
0/148 +0
—
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions' For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
Fraktur OCR
Recall 0%
Precision 0%
0/382 +0
—
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions' For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
Artefacts: Handschrift OCR – leicht – Antwort Handschrift OCR – leicht – Bild Handschrift OCR – mittel – Antwort Handschrift OCR – mittel – Bild Handschrift OCR – schwer – Antwort Handschrift OCR – schwer – Bild Fraktur OCR – Antwort Fraktur OCR – Bild

niah

all models in this bench →
Score
—
tok/s
0
Tokens
—
Wall
0.0 s
RAM
—
Worum geht es? Was wird getestet?
Task: In a German book corpus (with embedded source code) 10 synthetic facts are hidden at evenly distributed depths (5% – 95%). The model must retrieve all of them. Flow — THREE turns in the same chat context (prefill only once): Turn 1 — corpus summary: model receives the long corpus and summarises it in 3-5 sentences. Forces real processing of the text. Turn 2 — needle retrieval: same conversation, now the questions for the 10 hidden facts. Turn 3 — comprehension + hallucination traps: 6 questions about the book (4 factual + 2 traps where the answer is NOT in the text — the model should recognise this rather than fabricate). Default mode runs ONE uniform stage for all models: 120k tokens. Models without sufficient max_context are skipped at this stage. `niah_deep` additionally runs 32k / 64k / 200k for a full heatmap. Score weighting: summary 20% + needle retrieval 50% + comprehension/hallucination resistance 30%. Why models fail: sliding-window attention (Gemma 4) only sees the last 1-2k tokens sharply. Reasoning models hit the token limit before answering. Q4 KV cache measurably degrades recall at long contexts. On the hallucination traps the helpful bias lures models into plausible-sounding inventions.
Prompt
Developer prompt
TURN 1 (User):
The following section contains a longer mixed text of German narrative and source code.

===== TEXT BEGIN =====
<corpus with embedded needles, 32k–128k tokens depending on stage>
===== TEXT END =====

Summarise the text in 3-5 sentences. Mention the main characters, setting and key themes.

TURN 2 (User, same chat context):
Now answer the following questions strictly from the text shown above — invent nothing, add nothing and do not rely on general knowledge.

Questions:
1. <question for needle 1>
2. <question for needle 2>
...

Answer as a numbered list 1., 2., 3. with one short sentence each.

For each context length 10 needles are distributed across the corpus. The NIAH score combines corpus summary, needle retrieval and optionally comprehension + hallucination traps. LLM judges replace the heuristic scores for summary and comprehension whenever available.

Stage: 120k tokens
Sub-benchmark · needle retrieval Turn 2
0/10 hits
  • ○
    5% (5 %)
    expected: blauer Ankerstein, Lübeck-1907, A-318
  • ○
    10% (10 %)
    expected: smaragdgrün, Schlüssel, Seriennummer, 7-Bravo-12
  • ○
    15% (15 %)
    expected: Nordstern-Brigade, violetter
  • ○
    25% (25 %)
    expected: Indigo-Quark
  • ○
    33% (33 %)
    expected: Atlantis-Mira, NL-7711, 142
  • ○
    50% (50 %)
    expected: Ehrenmitglied, Aluminium-Gesellschaft, Köln
  • ○
    65% (65 %)
    expected: Safran-Klops Margarethe, 7 Gramm|7 g, 3 Stunden|drei Stunden
  • ○
    75% (75 %)
    expected: Erbe von Onkel Walpurgis, 42-Lima
  • ○
    85% (85 %)
    expected: 7f4a2e91-marlin-2026
  • ○
    90% (90 %)
    expected: Coriolis-Effekt, Pendel, Memo, Dr. Kühn, 17.03.2024
Model response to the needle questions 

              

                
Error: context-load failed: model konnte nicht mit 128000 Tokens geladen werden

            

          


            

              

                

                  Sub-benchmark · comprehension + hallucination traps
                  Turn 3
                

                

                    
                      0%
                    
                  
                    Facts 0/4 ·
                    Traps 0/2
                  
                

              

              
    
                  
  • 
                    
    
                        Factual question
                      
    Wie heißt der Schmied vom Blumental, bei dem Gottlieb in der Lehre ist? (Vor- und Nachname)
    
                    
    
                    
    
                  
    • 
                  
  • 
                    
    
                        Factual question
                      
    Wie heißt der französische Kapitän, der Gottliebs Eltern getötet hat?
    
                    
    
                    
    
                  
    • 
                  
  • 
                    
    
                        Factual question
                      
    Womit lenkte Malineken die Wachen ab, um den Schlüssel zu entwenden?
    
                    
    
                    
    
                  
    • 
                  
  • 
                    
    
                        Factual question
                      
    Warum muss Gottlieb sich verborgen halten? Was würden die Franzosen sonst tun?
    
                    
    
                    
    
                  
    • 
                  
  • 
                    
    
                        Hallucination trap
                      
    Wie heißt Gottliebs Onkel?
    
                    
    
                    
    
                  
    • 
                  
  • 
                    
    
                        Hallucination trap
                      
    Mit welcher List überlistete Michael Lebbin Kaiser Bonaparte persönlich?
    
                    
    
                    
    
                  
    • 
              

            

      



    

      Artefacts:
        Breakdown pro Kontextlänge
        Haystack 120k Tokens
    





  

    
tool_use

    all models in this bench →
  



  

    

      
Score

      

—      

    

    

      

        
tok/s

        
0

      

      

        
Tokens

        
—

      

      

        
Wall

        
0.0 s

      

      

        
RAM

        
—

      

    

  


    

      
        Worum geht es? Was wird getestet?
      
      
Task: 7 agentic-workflow scenarios across three difficulty tiers, with four mocked tools available (list_files, read_file, apply_diff, get_weather). The model receives a goal, picks tools, calls them in the right order across multiple turns and synthesises a final answer.

Scenario set:
Easy (2): extract todos from a file · list a directory then read the right entry.
Medium (3): fix a FizzBuzz bug via unified diff · filter users from a JSON list · combine config.json with a weather mock for an alert.
Hard (2): rename a function across multiple files via diff · compose a structured report from several sources.

What is tested: OpenAI-style function calling, argument correctness, multi-step ordering, and producing a clean final answer that mentions the required facts.

Why models fail: models without 'tool_use' capability ignore the schemas. Weak models pick wrong tools or emit malformed arguments. Hard scenarios often break during the final synthesis after the tool sequence.

    


    

      
        Prompt
      
      

          

            
System prompt

            
Du arbeitest in einer agentic IDE (vergleichbar mit Cursor/Claude Code). Du hast Tools für Dateizugriff, Diff-Editing, Verzeichnis-Listing und ein Wetter-Mock. Plane mehrstufig: erst aufklären, dann handeln, am Ende prüfen. Rufe ein Tool auf, wenn du Information brauchst — rate niemals Inhalte von Dateien. Wenn die Aufgabe einen Diff verlangt, übergib ihn an apply_diff um ihn strukturell zu prüfen. Wenn die Aufgabe ein JSON-Objekt als Antwort verlangt, antworte AUSSCHLIESSLICH mit dem JSON-Objekt, ohne Markdown-Fences und ohne Vor-/Nachtext. Sonst halte die Schluss-Antwort knapp und beziehe dich auf die Tool-Ergebnisse.

          

      

    


      

        

          

            easy_extract_todos
            leicht
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      

      

        

          

            easy_list_then_read
            leicht
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      

      

        

          

            medium_fizzbuzz_fix
            mittel
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      

      

        

          

            medium_filter_users
            mittel
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      

      

        

          

            medium_weather_alert
            mittel
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      

      

        

          

            hard_rename_function
            schwer
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      

      

        

          

            hard_compose_report
            schwer
          

            chat failed at iteration 0: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
        


        




        

          
Final answer

          

        

      



    

      Artefacts:
        Scenarios + Tool-Call-Verlauf
    





  

    
diagram_to_svg

    all models in this bench →
  



  

    

      
Score

      

—      

    

    

      

        
tok/s

        
0

      

      

        
Tokens

        
—

      

      

        
Wall

        
0.0 s

      

      

        
RAM

        
—

      

    

  


    

      
        Worum geht es? Was wird getestet?
      
      
Task: Photo of a hand-drawn diagram (architecture, sequence, quadrant matrix) → model must produce an inline-SVG representation of the same diagram.

Two score signals:
(1) Deterministic — SVG is parseable, has an <svg> root, enough elements and at least one <text>; all expected terms (boxes, labels) appear in the text content. Validity and term coverage each count for 15% of the final score.
(2) Qualitative — the `diagram-svg-judge` skill screenshots the SVG and visually compares it to the original along fixed axes (completeness, connections, arrow direction, grouping, layout readability, diagram-type fidelity, aesthetics). The judge counts 70%; aesthetics is double-weighted within the judge.

Why models fail: SVG generation requires spatial reasoning (positioning boxes, computing paths, setting viewBox) — noticeably harder than declarative Mermaid syntax. Weak VLMs often produce only an empty <svg> or an element salad without topology.

    


    

      
        Prompt
      
      

          

            
System prompt

            
Du bist Spezialist für Diagramm-Erkennung und SVG. Du gibst sauberes, parsbares SVG zurück, das jeder Browser ohne externe Ressourcen rendern kann.

          

          

            
Developer prompt

            
Auf dem Bild siehst du ein Diagramm (Architektur, Flowchart, Sequenz, Quadrant o.ä.). Erstelle eine SVG-Repräsentation des Diagramms.

Anforderungen:
- Antworte ausschließlich mit dem rohen SVG-Code, beginnend mit <svg ...> und endend mit </svg>. Keine Erklärungen, keine Markdown-Fences.
- Setze ein viewBox-Attribut (z.B. viewBox="0 0 1200 800"), damit das Bild skaliert.
- Nur Inline-Inhalt, keine externen Referenzen (kein <image href>, kein @import, kein xlink:href auf URLs).
- Alle im Diagramm sichtbaren Beschriftungen müssen als <text>-Elemente vorhanden und lesbar (Font-Size ≥ 12) sein.
- Verbindungen als <line>, <polyline> oder <path> mit deutlichem stroke. Pfeilspitzen via <marker>.
- Gruppiere zusammengehörige Teile mit <g>-Tags und sinnvollen id-Attributen.
- Wähle ausreichend Kontrast: dunkler Stroke auf weißem/hellem Hintergrund.
- Vermeide Überlappungen — plane das Layout so, dass Boxen nicht über Pfeilen liegen und Texte nicht aus ihren Boxen herausragen.
- Behalte die Struktur des Originals bei: Anzahl der Boxen, ihre Verbindungen und ihre Anordnung sollen vergleichbar sein.

          

      

    


      

        

          
diagram_eisenhower.png

          

              ✗ SVG not parseable
0%          

        

          
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400

      

      

        

          
diagram_service_architecture.png

          

              ✗ SVG not parseable
0%          

        

          
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400

      

      

        

          
diagram_sso_sequence.png

          

              ✗ SVG not parseable
0%          

        

          
Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400

      



    

      Artefacts:
        Source diagram_eisenhower
        SVG diagram_eisenhower
        Source diagram_service_architecture
        SVG diagram_service_architecture
        Source diagram_sso_sequence
        SVG diagram_sso_sequence
    





  

    
hallucination

    all models in this bench →
  



  

    

      
Score

      

50%
      

    

    

      

        
tok/s

        
0

      

      

        
Tokens

        
—

      

      

        
Wall

        
0.0 s

      

      

        
RAM

        
—

      

    

  


    

      
        Worum geht es? Was wird getestet?
      
      
Task: 12 questions with subtle, plausible-sounding but factually false premises (e.g. 'Which album did Tocotronic release in 1991?' — the band was only formed in 1993).

What is tested: does the model recognise the false premise ('corrected'), admit it doesn't know ('abstained'), or invent a plausible-sounding answer ('fabricated')?

Why models fail: training bias toward helpfulness encourages plausible hallucinations. Small models have weaker factual grounding. Subtle questions about personal details (e.g. politicians' children) are especially tempting to make up.

    


    

      
        Prompt
      
      

          

            
System prompt

            
Antworte knapp und faktisch korrekt auf Deutsch. Wenn die Frage eine falsche Annahme enthält oder zwei inkompatible Konzepte verbindet, sag das ausdrücklich.

          

      

    


    

        LLM-Judge aktiv
        (openai/gpt-5.4, Score 50%)
        · heuristische Klassifikation:
      0 korrigiert
      ·
      0 'weiß nicht'
      ·
      0 erfunden
      / 12 Fragen mit falscher Prämisse.
    

      

        

          

            
LLM-as-Judge

            

              korrekt 0 · ausgewichen 12 · falsch 0 · total 12
            

          

          

            50%
          

        

      

    
    
        
  • 
          
    
            
    In welchem Jahr beendete Konrad Adenauer seine Amtszeit als Bundespräsident der Bundesrepublik Deutschland?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    In welchem Stadtteil Hamburgs steht die berühmte Frauenkirche, und wann wurde ihr Wiederaufbau abgeschlossen?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Wie viele Zentimeter Mut passen in eine Schublade?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Welches Album veröffentlichte die Band Tocotronic im Jahr 1991?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    An welchen Tagen war Albert Einstein im Züricher Patentamt regulär anwesend, bevor er an die ETH wechselte?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Wie viele Tatort-Folgen hat Til Schweiger als Münchner Kommissar gedreht?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Wie viel Schlaf benötigt ein Vertrag pro Nacht?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    In welcher Tonart steht das Wort 'Donnerstag'?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Wie groß war die Wikingerflotte, die im 11. Jahrhundert von Lübeck aus operierte?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Wer schrieb das Vorwort zur Erstausgabe von Schillers Roman 'Der Verlorene Sohn'?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Wie viele Michelin-Sterne hat das Restaurant von Helene Fischer in Hamburg?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
        
  • 
          
    
            
    Bei welcher Luftfeuchtigkeit wachsen Wahrheiten am besten?
    
              
                J
              
          
    
          
    

    Fehler: Client error '400 Bad Request' for url 'http://127.0.0.1:1234/api/v0/chat/completions'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/400
            
    • 
    



    

      Artefacts:
        Alle Fragen + Antworten + Klassifikation