Model benchmark rankings

Compare capability scores, evaluation time, and reference cost across models and reasoning efforts.

Evaluated on private datasetsWhy we test this way

Data snapshot · 10/27/2026

Scoring dimension
About the scores

Interface behavior, user interactions, and state feedback. By default, each model shows its highest-scoring tested effort for the selected dimension.

Evaluation time and reference cost correspond to the selected dimension. They reflect benchmark runs, not everyday single-request latency or billing.

Explore evaluation tasksEvaluation method
Filters1
Result scope
46 models · 107 configsBest effort per model

Scroll horizontally to view all scores and costs →

RankModelEval. time
Speed tok/sActions
01
claude-opus-5-5
Backend engineering84Frontend interaction97Knowledge & Reasoning100Visual aesthetics73Maintainability70
89.158497100737048m 10s$12.11…
02
deepseek-v4.1-flash-1017NEW
Backend engineering72Frontend interaction96Knowledge & Reasoning75Visual aesthetics61Maintainability53
74.30729675615321m 0s$0.062…
02
gpt-6-astra
Backend engineering95Frontend interaction96Knowledge & Reasoning95Visual aesthetics75Maintainability61
88.80959695756116m 21s$1.51…
02
gpt-6.1-sol
Backend engineering95Frontend interaction96Knowledge & Reasoning90Visual aesthetics66Maintainability70
86.60959690667024m 27s$0.22…
05
gpt-6-sol
Backend engineering80Frontend interaction94Knowledge & Reasoning75Visual aesthetics75Maintainability66
78.9080947575666m 50s$0.22…
06
claude-fable-5-1
Backend engineering85Frontend interaction91Knowledge & Reasoning95Visual aesthetics80Maintainability68
87.25859195806826m 10s$6.68…
06
k3.1-proNEW
Backend engineering84Frontend interaction91Knowledge & Reasoning80Visual aesthetics68Maintainability64
79.60849180686421m 13s$0.7584
06
muse-spark-1.3
Backend engineering73Frontend interaction91Knowledge & Reasoning75Visual aesthetics59Maintainability47
72.6073917559473m 18s$0.014…
09
gpt-5.6-sol
Backend engineering78Frontend interaction90Knowledge & Reasoning65Visual aesthetics73Maintainability64
73.70789065736420m 30s$0.28…
10
claude-sonnet-5-5
Backend engineering86Frontend interaction89Knowledge & Reasoning85Visual aesthetics73Maintainability66
82.30868985736629m 45s$1.17…
11
claude-opus-4-8
Backend engineering64Frontend interaction88Knowledge & Reasoning85Visual aesthetics68Maintainability55
75.85648885685532m 18s$4.18…
11
deepseek-v4.1-flash
Backend engineering62Frontend interaction88Knowledge & Reasoning65Visual aesthetics64Maintainability55
67.8562886564558m 8s$0.057…
11
MiniMax-M3.1-FlashNEW
Backend engineering62Frontend interaction88Knowledge & Reasoning65Visual aesthetics64Maintainability51
67.4562886564519m 41s$0.024…
14
claude-opus-5
Backend engineering70Frontend interaction87Knowledge & Reasoning95Visual aesthetics73Maintainability61
81.70708795736124m 45s$3.20…
14
dotBotNEW
Backend engineering85Frontend interaction87Knowledge & Reasoning85Visual aesthetics70Maintainability73
81.9585878570732m 32sN/A…
14
Qwen 3.8 27b
Backend engineering64Frontend interaction87Knowledge & Reasoning20Visual aesthetics66Maintainability59
53.00648720665926m 31s$0.24…
14
Qwen 3.8 Flash
Backend engineering68Frontend interaction87Knowledge & Reasoning50Visual aesthetics73Maintainability57
65.15688750735729m 54s$0.050…
18
Gemini 3.8 Flash High
Backend engineering64Frontend interaction86Knowledge & Reasoning40Visual aesthetics66Maintainability59
59.8064864066593m 52s$0.062…
19
gpt-6.1-terraNEW
Backend engineering55Frontend interaction85Knowledge & Reasoning75Visual aesthetics47Maintainability59
67.2055857547591m 2s$0.072…
19
k3
Backend engineering70Frontend interaction85Knowledge & Reasoning70Visual aesthetics68Maintainability59
71.60708570685918m 4s$1.03…
19
mimo-v2.6-pro
Backend engineering70Frontend interaction85Knowledge & Reasoning65Visual aesthetics64Maintainability59
69.25708565645962m 35s$0.17…
19
muse-wonder
Backend engineering77Frontend interaction85Knowledge & Reasoning80Visual aesthetics73Maintainability57
77.0577858073573m 59s$0.72…
23
Gemini 4 Argon
Backend engineering82Frontend interaction84Knowledge & Reasoning85Visual aesthetics70Maintainability59
79.35828485705919m 55sN/A…
24
gpt-5.6-luna
Backend engineering56Frontend interaction83Knowledge & Reasoning25Visual aesthetics66Maintainability59
52.35568325665912m 59s$0.025…
25
glm-5.3
Backend engineering69Frontend interaction82Knowledge & Reasoning50Visual aesthetics66Maintainability68
64.40698250666823m 23s$0.43…
26
gpt-6-luna
Backend engineering62Frontend interaction80Knowledge & Reasoning50Visual aesthetics68Maintainability59
62.00628050685910m 29s$0.012…
26
grok-4.6
Backend engineering78Frontend interaction80Knowledge & Reasoning85Visual aesthetics64Maintainability59
76.85788085645931m 59s$0.40…
26
grok-4.7
Backend engineering87Frontend interaction80Knowledge & Reasoning85Visual aesthetics66Maintainability55
78.55878085665533m 52s$0.89…
26
space-bunny
Backend engineering61Frontend interaction80Knowledge & Reasoning75Visual aesthetics59Maintainability55
68.80618075595514m 30s$0.000…
30
claude-haiku-5-5
Backend engineering69Frontend interaction78Knowledge & Reasoning60Visual aesthetics59Maintainability64
65.6569786059648m 0s$0.066…
31
deepseek-v4-flash
Backend engineering58Frontend interaction76Knowledge & Reasoning55Visual aesthetics61Maintainability53
60.50587655615316m 48s$0.067…
31
glm-5.3-flash
Backend engineering55Frontend interaction76Knowledge & Reasoning20Visual aesthetics70Maintainability59
49.60557620705929m 43s$0.025…
31
gpt-5.6-terra
Backend engineering81Frontend interaction76Knowledge & Reasoning55Visual aesthetics61Maintainability61
65.90817655616123m 25s$0.38…
34
hy4-preview
Backend engineering74Frontend interaction73Knowledge & Reasoning65Visual aesthetics59Maintainability57
66.70747365595713m 25s$0.14…
35
k2.8
Backend engineering54Frontend interaction72Knowledge & Reasoning70Visual aesthetics57Maintainability53
63.55547270575323m 59s$0.25…
36
Gemini 3.7 Flash High
Backend engineering63Frontend interaction71Knowledge & Reasoning25Visual aesthetics70Maintainability59
51.9563712570594m 35s$0.058…
36
mimo-v2.6-flash
Backend engineering58Frontend interaction71Knowledge & Reasoning45Visual aesthetics59Maintainability57
56.10587145595775m 16s$0.068…
36
step-5
Backend engineering61Frontend interaction71Knowledge & Reasoning75Visual aesthetics47Maintainability55
65.20617175475520m 9s$0.24…
39
grok-botBotNEW
Backend engineering65Frontend interaction70Knowledge & Reasoning95Visual aesthetics68Maintainability59
76.3565709568593m 51sN/A…
grok-bot · Test resultsTime and reference cost correspond to Frontend interaction tasks.
Tested effortOverallBackend engineeringFrontend interactionKnowledge & ReasoningVisual aestheticsMaintainabilityEvaluation timeReference costSpeed tok/sCompare
UndisclosedShown in ranking
3m 51sN/A…

Scroll horizontally to view all scores and costs →

Score breakdown

Overall score
76.35
Evaluation time
39m 33s
Reference cost
N/A

Interface behavior, user interactions, and state feedback.

Layout and adaptation16 / 20

Content organization, layout constraints, and responsive adaptation.

Interaction behavior19 / 20

Input response, completed workflows, and edge-case interactions.

State continuity10 / 20

State propagation, recovery, and continuity across actions.

Accessibility and feedback10 / 15

Control semantics, accessibility support, and clear feedback.

State validation and regression15 / 25

Expected-state verification and behavioral regression checks.

40
Qwen 3.8 Max
Backend engineering64Frontend interaction69Knowledge & Reasoning40Visual aesthetics64Maintainability53
55.50646940645329m 19s$0.46…
41
doubao-seed-2-1-pro
Backend engineering68Frontend interaction68Knowledge & Reasoning55Visual aesthetics68Maintainability49
61.55686855684933m 16s$0.32…
42
MiniMax-M3
Backend engineering46Frontend interaction65Knowledge & Reasoning50Visual aesthetics59Maintainability44
52.95466550594414m 22s$0.14…
43
deepseek-v4-pro
Backend engineering67Frontend interaction63Knowledge & Reasoning65Visual aesthetics70Maintainability59
65.15676365705918m 41s$0.16…
44
MiniMax-M3.1-Flash-Preview
Backend engineering75Frontend interaction57Knowledge & Reasoning75Visual aesthetics70Maintainability53
68.45755775705312m 20sN/A…
45
deepseek-v4-flash-vision-exp
Backend engineering61Frontend interaction43Knowledge & Reasoning70Visual aesthetics61Maintainability55
59.95614370615511m 8s$0.045…
46
Nemotron-3-Ultra
Backend engineering46Frontend interaction40Knowledge & Reasoning40Visual aesthetics61Maintainability37
44.05464040613717m 46s$0.000…
Each model uses its highest-scoring effort here. Tied scores share a rank.
Interface behavior, user interactions, and state feedback. Evaluation scope and sources