Benchmark

Humanity’s Last Exam

Reasoning & knowledge

Categories
intelligence-index, reasoning, knowledge
Openness
unknown
Source
artificial_analysis
Reported scores
577

Reported scores

artificial_analysis

ModelOrganizationReported valueReportedEvidence
A.X-K2SK Telecom0.2956441149212232026-08-12evidence
Agnes 2.5 Pro AlphaSapiens AI0.335959221501392026-07-24evidence
Apertus 70B InstructSwiss AI Initiative0.05468025949953662025-09-02evidence
Apertus 8B InstructSwiss AI Initiative0.04958294717330862025-09-02evidence
Apriel-v1.5-15B-ThinkerServiceNow0.1214087117701582025-09-30evidence
Apriel-v1.6-15B-ThinkerServiceNow0.1079703429101022025-11-25evidence
Celeris-1Celeris0.06811862835959222026-07-24evidence
Claude 2.1Anthropic0.03892023-11-21evidence
Claude 3 HaikuAnthropic0.04077849860982392024-03-04evidence
Claude 3 OpusAnthropic0.02752024-03-04evidence
Claude 3 SonnetAnthropic0.03622024-03-04evidence
Claude 3.5 HaikuAnthropic0.03632024-10-22evidence
Claude 3.5 Sonnet (June '24)Anthropic0.03232024-06-21evidence
Claude 3.5 Sonnet (Oct '24)Anthropic0.03692024-10-22evidence
Claude 3.7 Sonnet (Non-reasoning)Anthropic0.04162025-02-24evidence
Claude 3.7 Sonnet (Reasoning)Anthropic0.09672025-02-24evidence
Claude 4 Opus (Non-reasoning)Anthropic0.0616311399443932025-05-22evidence
Claude 4 Opus (Reasoning)Anthropic0.1232622798887862025-05-22evidence
Claude 4 Sonnet (Non-reasoning)Anthropic0.04263206672845232025-05-22evidence
Claude 4 Sonnet (Reasoning)Anthropic0.1070435588507882025-05-22evidence
Claude 4.1 Opus (Reasoning)Anthropic0.1246524559777572025-08-05evidence
Claude 4.5 Haiku (Non-reasoning)Anthropic0.04216867469879522025-10-15evidence
Claude 4.5 Haiku (Reasoning)Anthropic0.1037998146431882025-10-15evidence
Claude 4.5 Sonnet (Non-reasoning)Anthropic0.0722891566265062025-09-29evidence
Claude 4.5 Sonnet (Reasoning)Anthropic0.1779425393883232025-09-29evidence
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic0.5546802594995372026-06-09evidence
Claude InstantAnthropic0.03532023-03-14evidence
Claude Opus 4.5 (Non-reasoning)Anthropic0.1316033364226142025-11-24evidence
Claude Opus 4.5 (Reasoning)Anthropic0.3012048192771082025-11-24evidence
Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic0.3994439295644112026-02-05evidence
Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic0.1909175162187212026-02-05evidence
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic0.4230769230769232026-04-16evidence
Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic0.3331788693234482026-04-16evidence
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic0.4865616311399442026-05-28evidence
Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic0.5282669138090832026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic0.4341983317886932026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic0.5486561631139942026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic0.5129749768303982026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic0.5440222428174242026-07-24evidence
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic0.335959221501392026-02-17evidence
Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic0.1334569045412422026-02-17evidence
Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic0.1121408711770162026-02-17evidence
Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic0.4128822984244672026-06-30evidence
Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic0.1904541241890642026-06-30evidence
Cogito v2.1 (Reasoning)Deep Cogito0.1200185356811862025-11-18evidence
Command ACohere0.04031510658016682025-03-13evidence
Command A+Cohere0.1195551436515292026-05-20evidence
Command-R (Mar '24)Cohere0.04792024-03-12evidence
Command-R+ (Apr '24)Cohere0.04632024-04-04evidence
DBRX InstructDatabricks0.02922024-03-27evidence
DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research0.04262025-02-13evidence
DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research0.03762025-03-13evidence
DeepSeek Coder V2 Lite InstructDeepSeek0.05362024-06-17evidence
DeepSeek R1 (Jan '25)DeepSeek0.0852025-01-20evidence
DeepSeek R1 0528 (May '25)DeepSeek0.1584800741427252025-05-28evidence
DeepSeek R1 0528 Qwen3 8BDeepSeek0.05931417979610752025-05-29evidence
DeepSeek R1 Distill Llama 70BDeepSeek0.05132025-01-20evidence
DeepSeek R1 Distill Qwen 1.5BDeepSeek0.03082025-01-20evidence
DeepSeek R1 Distill Qwen 14BDeepSeek0.04132025-01-20evidence
DeepSeek R1 Distill Qwen 32BDeepSeek0.04582025-01-20evidence
DeepSeek V3 (Dec '24)DeepSeek0.0292024-12-26evidence
DeepSeek V3 0324DeepSeek0.04742025-03-25evidence
DeepSeek V3.1 (Non-reasoning)DeepSeek0.06672845227062092025-08-21evidence
DeepSeek V3.1 (Reasoning)DeepSeek0.1427247451343842025-08-21evidence
DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek0.08665430954587582025-09-22evidence
DeepSeek V3.1 Terminus (Reasoning)DeepSeek0.1635773864689532025-09-22evidence
DeepSeek V3.2 (Non-reasoning)DeepSeek0.1121408711770162025-12-01evidence
DeepSeek V3.2 (Reasoning)DeepSeek0.2455977757182582025-12-01evidence
DeepSeek V3.2 Exp (Non-reasoning)DeepSeek0.09036144578313252025-09-29evidence
DeepSeek V3.2 Exp (Reasoning)DeepSeek0.1487488415199262025-09-29evidence
DeepSeek V3.2 SpecialeDeepSeek0.2873030583873962025-12-01evidence
DeepSeek V4 Flash (Non-reasoning)DeepSeek0.07784986098239112026-04-24evidence
DeepSeek V4 Flash (Reasoning, High Effort)DeepSeek0.302594995366082026-04-24evidence
DeepSeek V4 Flash (Reasoning, Max Effort)DeepSeek0.3484708063021322026-04-24evidence
DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek0.3855421686746992026-07-31evidence
DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek0.3447636700648752026-08-21evidence
DeepSeek V4 Pro (Non-reasoning)DeepSeek0.0824837812789622026-04-24evidence
DeepSeek V4 Pro (Reasoning, High Effort)DeepSeek0.3521779425393882026-04-24evidence
DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek0.3753475440222432026-04-24evidence
DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek0.4101019462465252026-08-13evidence
Devstral 2Mistral0.03568118628359592025-12-09evidence
Devstral MediumMistral0.03753475440222432025-07-10evidence
Devstral Small (Jul '25)Mistral0.03753475440222432025-07-10evidence
Devstral Small (May '25)Mistral0.04031510658016682025-05-21evidence
Devstral Small 2Mistral0.03475440222428172025-12-09evidence
DiffusionGemma 26B A4BGoogle0.1084337349397592026-06-10evidence
Doubao Seed CodeByteDance Seed0.1408711770157552025-11-11evidence
ERNIE 4.5 300B A47BBaidu0.03336422613531052025-06-30evidence
ERNIE 5.0 Thinking PreviewBaidu0.1325301204819282025-11-13evidence
Exaone 4.0 1.2B (Non-reasoning)LG AI Research0.05699721964782212025-07-15evidence
Exaone 4.0 1.2B (Reasoning)LG AI Research0.05977757182576462025-07-15evidence
EXAONE 4.0 32B (Non-reasoning)LG AI Research0.05004633920296572025-07-15evidence
EXAONE 4.0 32B (Reasoning)LG AI Research0.1144578313253012025-07-15evidence
EXAONE 4.5 33BLG AI Research0.1288229842446712026-04-09evidence
Falcon-H1R-7BTII UAE0.109823911028732026-01-04evidence
G9v3-39A5BAI9Stars0.1746987951807232026-08-20evidence
G9v3-3BAI9Stars0.04494902687673772026-07-23evidence
Gemini 1.0 ProGoogle0.04192023-12-06evidence
Gemini 1.5 Flash (May '24)Google0.04362024-05-14evidence
Gemini 1.5 Flash (Sep '24)Google0.03172024-09-24evidence
Gemini 1.5 Flash-8BGoogle0.04692024-10-03evidence
Gemini 1.5 Pro (May '24)Google0.03482024-05-15evidence
Gemini 1.5 Pro (Sep '24)Google0.04642024-09-24evidence
Gemini 2.0 Flash (experimental)Google0.04072024-12-11evidence
Gemini 2.0 Flash (Feb '25)Google0.04262025-02-05evidence
Gemini 2.0 Flash Thinking Experimental (Jan '25)Google0.06352025-01-21evidence
Gemini 2.0 Flash-Lite (Feb '25)Google0.03352025-02-25evidence
Gemini 2.0 Flash-Lite (Preview)Google0.04092025-02-05evidence
Gemini 2.0 Pro Experimental (Feb '25)Google0.06192025-02-05evidence
Gemini 2.5 Flash (Non-reasoning)Google0.04726598702502322025-05-20evidence
Gemini 2.5 Flash (Reasoning)Google0.1214087117701582025-05-20evidence
Gemini 2.5 Flash Preview (Non-reasoning)Google0.0382025-04-17evidence
Gemini 2.5 Flash Preview (Reasoning)Google0.1212025-04-17evidence
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google0.08711770157553292025-09-25evidence
Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google0.1380908248378132025-09-25evidence
Gemini 2.5 Flash-Lite (Non-reasoning)Google0.03660797034291012025-06-17evidence
Gemini 2.5 Flash-Lite (Reasoning)Google0.06811862835959222025-06-17evidence
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google0.05097312326227992025-09-25evidence
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google0.06950880444856352025-09-08evidence
Gemini 2.5 ProGoogle0.2252085264133462025-06-05evidence
Gemini 2.5 Pro Preview (Mar' 25)Google0.1802594995366082025-03-25evidence
Gemini 2.5 Pro Preview (May' 25)Google0.19932025-05-06evidence
Gemini 3 Flash Preview (Non-reasoning)Google0.149675625579242025-12-17evidence
Gemini 3 Flash Preview (Reasoning)Google0.3656163113994442025-12-17evidence
Gemini 3 Pro Preview (high)Google0.3971269694161262025-11-18evidence
Gemini 3 Pro Preview (low)Google0.2947173308619092025-11-18evidence
Gemini 3.1 Flash-LiteGoogle0.171918443002782026-03-03evidence
Gemini 3.1 Pro PreviewGoogle0.4703429101019462026-02-19evidence
Gemini 3.5 Flash (high)Google0.426784059314182026-05-19evidence
Gemini 3.5 Flash (medium)Google0.4133456904541242026-05-19evidence
Gemini 3.5 Flash (minimal)Google0.2409638554216872026-05-19evidence
Gemini 3.5 Flash-LiteGoogle0.1881371640407782026-07-21evidence
Gemini 3.6 Flash (high)Google0.4082483781278962026-07-21evidence
Gemini 3.7 Flash (high)Google0.4786839666357742026-08-13evidence
Gemini 3.7 Flash (low)Google0.3512511584800742026-08-13evidence
Gemini 3.7 Flash (medium)Google0.3897126969416132026-08-13evidence
Gemma 3 12B InstructGoogle0.04212025-03-12evidence
Gemma 3 1B InstructGoogle0.05329008341056532025-03-13evidence
Gemma 3 270MGoogle0.0361445783132532025-08-14evidence
Gemma 3 27B InstructGoogle0.0442025-03-12evidence
Gemma 3 4B InstructGoogle0.05292025-03-12evidence
Gemma 3n E2B InstructGoogle0.04170528266913812025-06-26evidence
Gemma 3n E4B InstructGoogle0.04494902687673772025-06-26evidence
Gemma 3n E4B Instruct Preview (May '25)Google0.04819277108433742025-05-20evidence
Gemma 4 12B (Non-reasoning)Google0.06348470806302132026-06-03evidence
Gemma 4 12B (Reasoning)Google0.1566265060240962026-06-03evidence
Gemma 4 26B A4B (Non-reasoning)Google0.1149212233549582026-04-02evidence
Gemma 4 26B A4B (Reasoning)Google0.1932344763670062026-04-02evidence
Gemma 4 31B (Non-reasoning)Google0.1181649675625582026-04-02evidence
Gemma 4 31B (Reasoning)Google0.2363299351251162026-04-02evidence
Gemma 4 E2B (Non-reasoning)Google0.04726598702502322026-04-02evidence
Gemma 4 E2B (Reasoning)Google0.04772937905468032026-04-02evidence
Gemma 4 E4B (Non-reasoning)Google0.04819277108433742026-04-03evidence
Gemma 4 E4B (Reasoning)Google0.03753475440222432026-04-03evidence
GLM 5V Turbo (Reasoning)Z.ai0.1714550509731232026-04-01evidence
GLM-4.5 (Reasoning)Z.ai0.1297497683039852025-07-28evidence
GLM-4.5-AirZ.ai0.07043558850787772025-07-28evidence
GLM-4.5V (Non-reasoning)Z.ai0.03475440222428172025-08-11evidence
GLM-4.5V (Reasoning)Z.ai0.06302131603336422025-08-11evidence
GLM-4.6 (Non-reasoning)Z.ai0.05468025949953662025-09-30evidence
GLM-4.6 (Reasoning)Z.ai0.1445783132530122025-09-30evidence
GLM-4.6V (Non-reasoning)Z.ai0.03660797034291012025-12-08evidence
GLM-4.6V (Reasoning)Z.ai0.09638554216867472025-12-08evidence
GLM-4.7 (Non-reasoning)Z.ai0.06394810009267842025-12-22evidence
GLM-4.7 (Reasoning)Z.ai0.273864689527342025-12-22evidence
GLM-4.7-Flash (Non-reasoning)Z.ai0.04958294717330862026-01-19evidence
GLM-4.7-Flash (Reasoning)Z.ai0.07599629286376272026-01-19evidence
GLM-5 (Non-reasoning)Z.ai0.07645968489341982026-02-11evidence
GLM-5 (Reasoning)Z.ai0.2928637627432812026-02-11evidence
GLM-5-TurboZ.ai0.2775718257645972026-03-15evidence
GLM-5.1 (Non-reasoning)Z.ai0.2794253938832252026-04-07evidence
GLM-5.1 (Reasoning)Z.ai0.3007414272474512026-04-07evidence
GLM-5.2 (max)Z.ai0.4114921223354962026-06-16evidence
GLM-5.2 (Non-reasoning)Z.ai0.0977757182576462026-06-16evidence
GLM-5.3 (max)Z.ai0.4226135310472662026-08-18evidence
GPT-4 TurboOpenAI0.03132023-11-06evidence
GPT-4.1OpenAI0.04182025-04-14evidence
GPT-4.1 miniOpenAI0.05022025-04-14evidence
GPT-4.1 nanoOpenAI0.03753475440222432025-04-14evidence
GPT-4o (Aug '24)OpenAI0.02332024-08-06evidence
GPT-4o (ChatGPT)OpenAI0.02732025-02-15evidence
GPT-4o (March 2025, chatgpt-4o-latest)OpenAI0.03972025-03-27evidence
GPT-4o (May '24)OpenAI0.01762024-05-13evidence
GPT-4o (Nov '24)OpenAI0.02422024-11-20evidence
GPT-4o miniOpenAI0.0422024-07-18evidence
GPT-5 (ChatGPT)OpenAI0.06626506024096392025-08-07evidence
GPT-5 (high)OpenAI0.284986098239112025-08-07evidence
GPT-5 (low)OpenAI0.1955514365152922025-08-07evidence
GPT-5 (medium)OpenAI0.2539388322520852025-08-07evidence
GPT-5 (minimal)OpenAI0.06024096385542172025-08-07evidence
GPT-5 Codex (high)OpenAI0.2784986098239112025-09-23evidence
GPT-5 mini (high)OpenAI0.2145505097312332025-08-07evidence
GPT-5 mini (medium)OpenAI0.1594068582020392025-08-07evidence
GPT-5 mini (minimal)OpenAI0.05050973123262282025-08-07evidence
GPT-5 nano (high)OpenAI0.09499536607970342025-08-07evidence
GPT-5 nano (medium)OpenAI0.08711770157553292025-08-07evidence
GPT-5 nano (minimal)OpenAI0.03985171455050972025-08-07evidence
GPT-5.1 (high)OpenAI0.284986098239112025-11-13evidence
GPT-5.1 (Non-reasoning)OpenAI0.05282669138090832025-11-13evidence
GPT-5.1 Codex (high)OpenAI0.2571825764596852025-11-13evidence
GPT-5.1 Codex mini (high)OpenAI0.1848934198331792025-11-13evidence
GPT-5.2 (medium)OpenAI0.2669138090824842025-12-11evidence
GPT-5.2 (Non-reasoning)OpenAI0.08016682113067662025-12-11evidence
GPT-5.2 (xhigh)OpenAI0.3767377201112142025-12-11evidence
GPT-5.2 Codex (xhigh)OpenAI0.3572752548656162025-12-11evidence
GPT-5.3 Codex (xhigh)OpenAI0.4249304911955512026-02-05evidence
GPT-5.4 (low)OpenAI0.3081556997219652026-03-05evidence
GPT-5.4 (Non-reasoning)OpenAI0.1126042632066732026-03-05evidence
GPT-5.4 (xhigh)OpenAI0.4374420759962932026-03-05evidence
GPT-5.4 mini (medium)OpenAI0.1858202038924932026-03-17evidence
GPT-5.4 mini (Non-Reasoning)OpenAI0.05885078776645042026-03-17evidence
GPT-5.4 mini (xhigh)OpenAI0.2812789620018542026-03-17evidence
GPT-5.4 nano (medium)OpenAI0.1594068582020392026-03-17evidence
GPT-5.4 nano (Non-Reasoning)OpenAI0.04077849860982392026-03-17evidence
GPT-5.4 nano (xhigh)OpenAI0.2826691380908252026-03-17evidence
GPT-5.5 (high)OpenAI0.4504170528266912026-04-23evidence
GPT-5.5 (low)OpenAI0.3266913809082482026-04-23evidence
GPT-5.5 (medium)OpenAI0.423540315106582026-04-23evidence
GPT-5.5 (Non-reasoning)OpenAI0.1371640407784992026-04-23evidence
GPT-5.5 (xhigh)OpenAI0.4578313253012052026-04-23evidence
GPT-5.5 Instant (June 2026)OpenAI0.1992585727525492026-06-25evidence
GPT-5.5 Instant (May 2026)OpenAI0.2159406858202042026-05-05evidence
GPT-5.6 Luna (high)OpenAI0.3341056533827622026-07-09evidence
GPT-5.6 Luna (low)OpenAI0.1983317886932342026-07-09evidence
GPT-5.6 Luna (max)OpenAI0.3948100092678412026-07-09evidence
GPT-5.6 Luna (medium)OpenAI0.2576459684893422026-07-09evidence
GPT-5.6 Luna (Non-reasoning)OpenAI0.0722891566265062026-07-09evidence
GPT-5.6 Luna (xhigh)OpenAI0.3697868396663582026-07-09evidence
GPT-5.6 Sol (high)OpenAI0.460148285449492026-07-09evidence
GPT-5.6 Sol (low)OpenAI0.3938832252085262026-07-09evidence
GPT-5.6 Sol (max)OpenAI0.4949026876737722026-07-09evidence
GPT-5.6 Sol (medium)OpenAI0.4221501390176092026-07-09evidence
GPT-5.6 Sol (Non-reasoning)OpenAI0.1668211306765522026-07-09evidence
GPT-5.6 Sol (xhigh)OpenAI0.4731232622798892026-07-09evidence
GPT-5.6 Terra (high)OpenAI0.3850787766450422026-07-09evidence
GPT-5.6 Terra (low)OpenAI0.2919369786839672026-07-09evidence
GPT-5.6 Terra (max)OpenAI0.4291010194624652026-07-09evidence
GPT-5.6 Terra (medium)OpenAI0.3327154772937912026-07-09evidence
GPT-5.6 Terra (Non-reasoning)OpenAI0.1139944392956442026-07-09evidence
GPT-5.6 Terra (xhigh)OpenAI0.4189063948100092026-07-09evidence
gpt-oss-120b (high)OpenAI0.1960148285449492025-08-05evidence
gpt-oss-120b (low)OpenAI0.05885078776645042025-08-05evidence
gpt-oss-20b (high)OpenAI0.109823911028732025-08-05evidence
gpt-oss-20b (low)OpenAI0.05282669138090832025-08-05evidence
Granite 3.3 8B (Non-reasoning)IBM0.04170528266913812025-04-16evidence
Granite 4.0 1BIBM0.04819277108433742025-10-28evidence
Granite 4.0 350MIBM0.05514365152919372025-10-28evidence
Granite 4.0 H 1BIBM0.05004633920296572025-10-28evidence
Granite 4.0 H 350MIBM0.06441149212233552025-10-28evidence
Granite 4.0 H SmallIBM0.03799814643188142025-09-22evidence
Granite 4.0 MicroIBM0.05004633920296572025-09-22evidence
Granite 4.1 30BIBM0.0412418906394812026-04-29evidence
Granite 4.1 3BIBM0.03429101019462472026-04-29evidence
Granite 4.1 8BIBM0.03799814643188142026-04-29evidence
Grok 2 (Dec '24)xAI0.03062024-12-12evidence
Grok 3xAI0.04122025-02-19evidence
Grok 3 mini Reasoning (high)xAI0.11012025-02-19evidence
Grok 4xAI0.2669138090824842025-07-10evidence
Grok 4 Fast (Non-reasoning)xAI0.04541241890639482025-09-19evidence
Grok 4 Fast (Reasoning)xAI0.1909175162187212025-09-19evidence
Grok 4.1 Fast (Non-reasoning)xAI0.05097312326227992025-11-19evidence
Grok 4.1 Fast (Reasoning)xAI0.1932344763670062025-11-19evidence
Grok 4.20 0309 (Non-reasoning)xAI0.2451343836886012026-03-10evidence
Grok 4.20 0309 (Reasoning)xAI0.3239110287303062026-03-10evidence
Grok 4.20 0309 v2 (Non-reasoning)xAI0.2794253938832252026-04-07evidence
Grok 4.20 0309 v2 (Reasoning)xAI0.3452270620945322026-04-07evidence
Grok 4.3 (high)xAI0.3721037998146432026-04-30evidence
Grok 4.3 (low)xAI0.1835032437442082026-04-30evidence
Grok 4.3 (medium)xAI0.2998146431881372026-04-30evidence
Grok 4.3 (Non-reasoning)xAI0.06765523632993512026-04-30evidence
Grok 4.5 (high)xAI0.426784059314182026-07-08evidence
Grok 4.6 (high)xAI0.4291010194624652026-08-12evidence
Grok 4.6 (low)xAI0.2761816496756262026-08-12evidence
Grok 4.6 (medium)xAI0.4212233549582952026-08-12evidence
Grok 4.6 (xhigh)xAI0.4406858202038932026-08-12evidence
Grok BetaxAI0.04452024-08-13evidence
Grok Build 0.1 0616xAI0.3827618164967562026-06-16evidence
Grok Code Fast 1xAI0.08016682113067662025-08-28evidence
Hermes 3 - Llama-3.1 70BNous Research0.03992024-08-15evidence
Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research0.04216867469879522025-08-27evidence
Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research0.1088971269694162025-08-27evidence
Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research0.03568118628359592025-08-27evidence
Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research0.087581093605192025-08-27evidence
Hy3Tencent0.3345690454124192026-07-06evidence
Hy3-preview (Non-reasoning)Tencent0.06950880444856352026-04-23evidence
Hy3-preview (Reasoning)Tencent0.2775718257645972026-04-23evidence
HyperCLOVA X SEED Think (32B)Naver0.05468025949953662025-12-26evidence
HyperNova 60B 2605Multiverse Computing0.1663577386468952026-05-26evidence
Inkling (xhigh)Thinking Machines0.3188137164040782026-07-15evidence
Inkling SmallThinking Machines0.3331788693234482026-07-30evidence
INTELLECT-3Prime Intellect0.1306765523632992025-11-27evidence
Jamba 1.5 LargeAI21 Labs0.04072024-08-22evidence
Jamba 1.5 MiniAI21 Labs0.05142024-08-22evidence
Jamba 1.6 LargeAI21 Labs0.03612025-03-06evidence
Jamba 1.6 MiniAI21 Labs0.04312025-03-06evidence
Jamba 1.7 LargeAI21 Labs0.03660797034291012025-07-07evidence
Jamba 1.7 MiniAI21 Labs0.04448563484708062025-07-07evidence
Jamba Reasoning 3BAI21 Labs0.03753475440222432025-10-08evidence
JT-35B-FlashChina Mobile0.06441149212233552026-05-14evidence
JT-4.1 Flash 236B A21BChina Mobile0.1779425393883232026-07-09evidence
JT-MINIChina Mobile0.06441149212233552026-04-15evidence
K-EXAONE (Non-reasoning)LG AI Research0.05699721964782212025-12-31evidence
K-EXAONE (Reasoning)LG AI Research0.1394810009267842025-12-31evidence
K-EXAONE 2.0 0803LG AI Research0.1858202038924932026-08-12evidence
K2 Think V2MBZUAI Institute of Foundation Models0.1010194624652462025-12-15evidence
K2-V2 (high)MBZUAI Institute of Foundation Models0.1051899907321592025-12-05evidence
K2-V2 (low)MBZUAI Institute of Foundation Models0.0361445783132532025-12-05evidence
K2-V2 (medium)MBZUAI Institute of Foundation Models0.04355885078776652025-12-05evidence
KAT Coder Pro V2KwaiKAT0.1612604263206672026-03-27evidence
KAT-Coder-Pro V1KwaiKAT0.3364226135310472025-11-11evidence
Kimi K2Moonshot AI0.07414272474513442025-07-11evidence
Kimi K2 0905Moonshot AI0.06394810009267842025-09-05evidence
Kimi K2 ThinkingMoonshot AI0.2381835032437442025-11-06evidence
Kimi K2.5 (Non-reasoning)Moonshot AI0.1316033364226142026-01-27evidence
Kimi K2.5 (Reasoning)Moonshot AI0.3072289156626512026-01-27evidence
Kimi K2.6Moonshot AI0.3748841519925862026-04-20evidence
Kimi K2.6 (Non-reasoning)Moonshot AI0.1955514365152922026-04-20evidence
Kimi K2.7 CodeMoonshot AI0.350324374420762026-06-12evidence
Kimi K3 (low)Moonshot AI0.2497683039851712026-07-16evidence
Kimi K3 (max)Moonshot AI0.4689527340129752026-07-16evidence
Kimi Linear 48B A3B InstructMoonshot AI0.02455977757182582025-10-30evidence
LFM 40BLiquid AI0.04932024-09-30evidence
LFM2 1.2BLiquid AI0.05560704355885082025-07-10evidence
LFM2 2.6BLiquid AI0.05468025949953662025-09-23evidence
LFM2 24B A2BLiquid AI0.04216867469879522026-02-25evidence
LFM2 8B A1BLiquid AI0.04911955514365152025-10-07evidence
LFM2.5-1.2B-InstructLiquid AI0.0671918443002782026-01-05evidence
LFM2.5-1.2B-ThinkingLiquid AI0.062094531974052026-01-20evidence
LFM2.5-2.6BLiquid AI0.062094531974052026-08-04evidence
LFM2.5-8B-A1BLiquid AI0.06858202038924932026-05-28evidence
LFM2.5-VL-1.6BLiquid AI0.05097312326227992026-01-05evidence
Ling 2.6 FlashInclusionAI0.06255792400370712026-04-21evidence
Ling 3.0 FlashInclusionAI0.2367933271547732026-08-04evidence
Ling 3.0 TinyInclusionAI0.09314179796107512026-08-06evidence
Ling-1TInclusionAI0.07321594068582022025-10-08evidence
Ling-2.6-1TInclusionAI0.08711770157553292026-04-23evidence
Ling-flash-2.0InclusionAI0.062094531974052025-09-17evidence
Ling-mini-2.0InclusionAI0.05050973123262282025-09-09evidence
Llama 2 Chat 13BMeta0.04772023-07-18evidence
Llama 2 Chat 70BMeta0.05212023-07-18evidence
Llama 2 Chat 7BMeta0.05332023-07-18evidence
Llama 3 Instruct 70BMeta0.04522024-04-18evidence
Llama 3 Instruct 8BMeta0.05062024-04-18evidence
Llama 3.1 Instruct 405BMeta0.03982024-07-23evidence
Llama 3.1 Instruct 70BMeta0.04472024-07-23evidence
Llama 3.1 Instruct 8BMeta0.05292024-07-23evidence
Llama 3.1 Nemotron Instruct 70BNVIDIA0.04192024-10-15evidence
Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIA0.05189990732159412025-05-20evidence
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA0.07382025-04-07evidence
Llama 3.1 Tulu3 405BAllen Institute for AI0.03282025-01-30evidence
Llama 3.2 Instruct 11B (Vision)Meta0.05532024-09-25evidence
Llama 3.2 Instruct 1BMeta0.05462024-09-25evidence
Llama 3.2 Instruct 3BMeta0.05352024-09-25evidence
Llama 3.2 Instruct 90B (Vision)Meta0.04492024-09-25evidence
Llama 3.3 Instruct 70BMeta0.03562024-12-06evidence
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA0.03812025-03-18evidence
Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIA0.05622025-03-18evidence
Llama 4 MaverickMeta0.04911955514365152025-04-05evidence
Llama 4 ScoutMeta0.03782025-04-05evidence
Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA0.04309545875810942025-07-25evidence
Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA0.07321594068582022025-07-25evidence
LongCat 2.0LongCat0.3368860055607042026-06-29evidence
LongCat Flash LiteLongCat0.05838739573679332026-01-28evidence
Magistral Medium 1Mistral0.09823911028730312025-06-10evidence
Magistral Medium 1.2Mistral0.1028730305838742025-09-18evidence
Magistral Small 1Mistral0.07553290083410572025-06-10evidence
Magistral Small 1.2Mistral0.06441149212233552025-09-17evidence
Mercury 2Inception0.1714550509731232026-02-20evidence
Mi:dm K 2.5 ProKorea Telecom0.08063021316033362025-12-11evidence
Mi:dm K 2.5 Pro PreviewKorea Telecom0.09082483781278962025-12-11evidence
MiMo-V2-Flash (Feb 2026)Xiaomi0.2210379981464322025-12-16evidence
MiMo-V2-Flash (Non-reasoning)Xiaomi0.08572752548656162025-12-16evidence
MiMo-V2-Flash (Reasoning)Xiaomi0.2284522706209452025-12-16evidence
MiMo-V2-OmniXiaomi0.2210379981464322026-03-19evidence
MiMo-V2-Omni-0327Xiaomi0.2256719184430032026-03-27evidence
MiMo-V2-ProXiaomi0.3044485634847082026-03-18evidence
MiMo-V2.5Xiaomi0.2720111214087122026-04-22evidence
MiMo-V2.5-ProXiaomi0.3568118628359592026-04-22evidence
MiMo-V2.5-Pro (Non-reasoning)Xiaomi0.1478220574606122026-04-22evidence
MiniCPM-V 4.6 1.3BOpenBMB0.05050973123262282026-05-11evidence
MiniCPM5-1B (Non-reasoning)OpenBMB0.04541241890639482026-05-25evidence
MiniCPM5-1B (Reasoning)OpenBMB0.06487488415199262026-05-25evidence
MiniMax M1 40kMiniMax0.07831325301204822025-06-17evidence
MiniMax M1 80kMiniMax0.08897126969416132025-06-17evidence
MiniMax-M2MiniMax0.1367006487488422025-10-26evidence
MiniMax-M2.1MiniMax0.2316960148285452025-12-23evidence
MiniMax-M2.5MiniMax0.2052826691380912026-02-12evidence
MiniMax-M2.7MiniMax0.296107506950882026-03-18evidence
MiniMax-M3MiniMax0.3897126969416132026-06-01evidence
Ministral 3 14BMistral0.0463392029657092025-12-02evidence
Ministral 3 3BMistral0.05375347544022242025-12-02evidence
Ministral 3 8BMistral0.04263206672845232025-12-02evidence
Mistral 7B InstructMistral0.04552023-09-27evidence
Mistral Large (Feb '24)Mistral0.0352024-02-26evidence
Mistral Large 2 (Jul '24)Mistral0.02952024-07-24evidence
Mistral Large 2 (Nov '24)Mistral0.03342024-11-18evidence
Mistral Large 3Mistral0.04170528266913812025-12-02evidence
Mistral MediumMistral0.03512023-12-11evidence
Mistral Medium 3Mistral0.04052025-05-07evidence
Mistral Medium 3.1Mistral0.04680259499536612025-08-12evidence
Mistral Medium 3.5Mistral0.1376274328081562026-04-29evidence
Mistral SabaMistral0.0432025-02-17evidence
Mistral Small (Feb '24)Mistral0.04092024-02-26evidence
Mistral Small (Sep '24)Mistral0.04252024-09-17evidence
Mistral Small 3Mistral0.03762025-01-30evidence
Mistral Small 3.1Mistral0.04309545875810942025-03-17evidence
Mistral Small 3.2Mistral0.04309545875810942025-06-20evidence
Mistral Small 4 (Non-reasoning)Mistral0.03799814643188142026-03-16evidence
Mistral Small 4 (Reasoning)Mistral0.09870250231696022026-03-16evidence
Mixtral 8x22B InstructMistral0.04012024-04-17evidence
Mixtral 8x7B InstructMistral0.04682023-12-11evidence
Molmo 7B-DAllen Institute for AI0.0514365152919372024-09-25evidence
Molmo2-8BAllen Institute for AI0.04263206672845232025-12-11evidence
Motif 3Motif Technologies0.3697868396663582026-08-12evidence
Motif 3 (Beta)Motif Technologies0.4036144578313252026-07-14evidence
Motif-2-12.7B-ReasoningMotif Technologies0.087581093605192025-12-04evidence
Muse Glimmer (high)Meta0.2196478220574612026-08-10evidence
Muse SparkMeta0.4068582020389252026-04-08evidence
Muse Spark 1.1 (xhigh)Meta0.4620018535681192026-07-09evidence
Muse Spark 1.2 (xhigh)Meta0.4545875810936052026-08-05evidence
Nanbeige4.1-3BNanbeige0.1088971269694162026-02-11evidence
Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA0.04819277108433742026-04-29evidence
Nemotron 3 Super 120B A12B (Reasoning)NVIDIA0.2075996292863762026-03-11evidence
Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA0.2840593141797962026-06-04evidence
Nemotron 3.5 LightningNVIDIA0.1056533827618172026-08-11evidence
Nemotron Cascade 2 30B A3BNVIDIA0.1200185356811862026-03-19evidence
Nex-N2-ProNex AGI0.3373493975903612026-06-02evidence
North Mini CodeCohere0.1107506950880442026-06-09evidence
Nova 2.0 Lite (high)Amazon0.116311399443932025-10-29evidence
Nova 2.0 Lite (low)Amazon0.04031510658016682025-10-29evidence
Nova 2.0 Lite (medium)Amazon0.08989805375347542025-10-29evidence
Nova 2.0 Lite (Non-reasoning)Amazon0.02873030583873962025-10-29evidence
Nova 2.0 Omni (medium)Amazon0.06950880444856352025-11-26evidence
Nova 2.0 Omni (Non-reasoning)Amazon0.03846153846153852025-11-26evidence
Nova 2.0 Pro Preview (low)Amazon0.05189990732159412025-11-27evidence
Nova 2.0 Pro Preview (medium)Amazon0.09406858202038922025-11-27evidence
Nova 2.0 Pro Preview (Non-reasoning)Amazon0.03938832252085262025-11-27evidence
Nova LiteAmazon0.04292024-12-03evidence
Nova MicroAmazon0.0463392029657092024-12-03evidence
Nova PremierAmazon0.04170528266913812025-04-30evidence
Nova ProAmazon0.03192024-12-03evidence
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA0.04587581093605192025-12-15evidence
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA0.1139944392956442025-12-15evidence
NVIDIA Nemotron 3 Nano 4BNVIDIA0.04865616311399442026-03-16evidence
NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA0.04263206672845232025-10-28evidence
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA0.05514365152919372025-10-28evidence
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA0.04587581093605192025-08-18evidence
NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA0.04865616311399442025-08-18evidence
o1OpenAI0.07012024-12-05evidence
o1-miniOpenAI0.03572024-09-12evidence
o3OpenAI0.20052025-04-16evidence
o3-miniOpenAI0.07912025-01-31evidence
o3-mini (high)OpenAI0.12042025-01-31evidence
o4-mini (high)OpenAI0.1654309545875812025-04-16evidence
OLMo 2 32BAllen Institute for AI0.0361445783132532025-03-13evidence
OLMo 2 7BAllen Institute for AI0.05375347544022242024-11-26evidence
Olmo 3 32B ThinkAllen Institute for AI0.06441149212233552025-11-20evidence
Olmo 3 7B InstructAllen Institute for AI0.05792400370713622025-11-20evidence
Olmo 3 7B ThinkAllen Institute for AI0.05977757182576462025-11-20evidence
Olmo 3.1 32B InstructAllen Institute for AI0.05004633920296572026-01-13evidence
Olmo 3.1 32B ThinkAllen Institute for AI0.06255792400370712025-12-12evidence
OpenChat 3.5 (1210)OpenChat0.04792023-12-18evidence
Phi-3 Mini Instruct 3.8BMicrosoft0.04962024-04-23evidence
Phi-4Microsoft0.03762024-12-12evidence
Phi-4 Mini InstructMicrosoft0.04452024-02-26evidence
Phi-4 Multimodal InstructMicrosoft0.052025-02-26evidence
Pixtral LargeMistral0.0282024-11-18evidence
Qwen2 Instruct 72BQwen0.03662024-06-07evidence
Qwen2.5 Coder Instruct 32BQwen0.03512024-11-11evidence
Qwen2.5 Coder Instruct 7BQwen0.04892024-09-19evidence
Qwen2.5 Instruct 32BQwen0.03992024-09-19evidence
Qwen2.5 Instruct 72BQwen0.03562024-09-19evidence
Qwen2.5 MaxQwen0.03822025-01-28evidence
Qwen2.5 TurboQwen0.04092024-11-18evidence
Qwen3 0.6B (Non-reasoning)Qwen0.04932025-04-28evidence
Qwen3 0.6B (Reasoning)Qwen0.05642025-04-28evidence
Qwen3 1.7B (Non-reasoning)Qwen0.05322025-04-28evidence
Qwen3 1.7B (Reasoning)Qwen0.04642025-04-28evidence
Qwen3 14B (Non-reasoning)Qwen0.04092025-04-28evidence
Qwen3 14B (Reasoning)Qwen0.04532025-04-28evidence
Qwen3 235B A22B (Non-reasoning)Qwen0.04192025-04-28evidence
Qwen3 235B A22B (Reasoning)Qwen0.10982025-04-28evidence
Qwen3 235B A22B 2507 (Reasoning)Qwen0.1589434661723822025-07-25evidence
Qwen3 235B A22B 2507 InstructQwen0.1112140871177022025-07-21evidence
Qwen3 30B A3B (Non-reasoning)Qwen0.0462025-04-28evidence
Qwen3 30B A3B (Reasoning)Qwen0.06162025-04-28evidence
Qwen3 30B A3B 2507 (Reasoning)Qwen0.1028730305838742025-07-30evidence
Qwen3 30B A3B 2507 InstructQwen0.06904541241890642025-07-29evidence
Qwen3 32B (Non-reasoning)Qwen0.04132025-04-28evidence
Qwen3 32B (Reasoning)Qwen0.07412025-04-28evidence
Qwen3 4B (Non-reasoning)Qwen0.03322025-04-28evidence
Qwen3 4B (Reasoning)Qwen0.04422025-04-28evidence
Qwen3 4B 2507 (Reasoning)Qwen0.0616311399443932025-08-06evidence
Qwen3 4B 2507 InstructQwen0.04494902687673772025-08-06evidence
Qwen3 8B (Non-reasoning)Qwen0.01952025-04-28evidence
Qwen3 8B (Reasoning)Qwen0.03892025-04-28evidence
Qwen3 Coder 30B A3B InstructQwen0.03846153846153852025-07-31evidence
Qwen3 Coder 480B A35B InstructQwen0.04541241890639482025-07-22evidence
Qwen3 Coder NextQwen0.1014828544949032026-02-03evidence
Qwen3 MaxQwen0.1190917516218722025-09-23evidence
Qwen3 Max (Preview)Qwen0.1014828544949032025-09-05evidence
Qwen3 Max ThinkingQwen0.2798887859128822026-01-26evidence
Qwen3 Max Thinking (Preview)Qwen0.1265060240963862025-11-03evidence
Qwen3 Next 80B A3B (Reasoning)Qwen0.1255792400370712025-09-11evidence
Qwen3 Next 80B A3B InstructQwen0.07645968489341982025-09-11evidence
Qwen3 Omni 30B A3B (Reasoning)Qwen0.07460611677479152025-09-22evidence
Qwen3 Omni 30B A3B InstructQwen0.0463392029657092025-09-22evidence
Qwen3 VL 235B A22B (Reasoning)Qwen0.1190917516218722025-09-23evidence
Qwen3 VL 235B A22B InstructQwen0.06626506024096392025-09-23evidence
Qwen3 VL 30B A3B (Reasoning)Qwen0.08943466172381842025-10-03evidence
Qwen3 VL 30B A3B InstructQwen0.06302131603336422025-10-03evidence
Qwen3 VL 32B (Reasoning)Qwen0.1014828544949032025-10-21evidence
Qwen3 VL 32B InstructQwen0.06811862835959222025-10-21evidence
Qwen3 VL 4B (Reasoning)Qwen0.04587581093605192025-10-14evidence
Qwen3 VL 4B InstructQwen0.0361445783132532025-10-14evidence
Qwen3 VL 8B (Reasoning)Qwen0.03846153846153852025-10-14evidence
Qwen3 VL 8B InstructQwen0.02687673772011122025-10-14evidence
Qwen3.5 0.8B (Non-reasoning)Qwen0.05097312326227992026-03-02evidence
Qwen3.5 0.8B (Reasoning)Qwen0.01065801668211312026-03-02evidence
Qwen3.5 122B A10B (Non-reasoning)Qwen0.1594068582020392026-02-24evidence
Qwen3.5 122B A10B (Reasoning)Qwen0.2520852641334572026-02-24evidence
Qwen3.5 27B (Non-reasoning)Qwen0.1394810009267842026-02-24evidence
Qwen3.5 27B (Reasoning)Qwen0.2391102873030582026-02-24evidence
Qwen3.5 2B (Non-reasoning)Qwen0.04958294717330862026-03-02evidence
Qwen3.5 2B (Reasoning)Qwen0.0259499536607972026-03-02evidence
Qwen3.5 35B A3B (Non-reasoning)Qwen0.1343836886005562026-02-24evidence
Qwen3.5 35B A3B (Reasoning)Qwen0.2103799814643192026-02-24evidence
Qwen3.5 397B A17B (Non-reasoning)Qwen0.1983317886932342026-02-16evidence
Qwen3.5 397B A17B (Reasoning)Qwen0.2896200185356812026-02-16evidence
Qwen3.5 4B (Non-reasoning)Qwen0.07970342910101952026-03-02evidence
Qwen3.5 4B (Reasoning)Qwen0.09916589434661722026-03-02evidence
Qwen3.5 9B (Non-reasoning)Qwen0.09360518999073222026-03-02evidence
Qwen3.5 9B (Reasoning)Qwen0.1492122335495832026-03-02evidence
Qwen3.5 Omni FlashQwen0.07599629286376272026-03-30evidence
Qwen3.5 Omni PlusQwen0.1487488415199262026-03-30evidence
Qwen3.6 27B (Non-reasoning)Qwen0.1506024096385542026-04-22evidence
Qwen3.6 27B (Reasoning)Qwen0.2307692307692312026-04-22evidence
Qwen3.6 35B A3B (Non-reasoning)Qwen0.1390176088971272026-04-16evidence
Qwen3.6 35B A3B (Reasoning)Qwen0.2224281742354032026-04-16evidence
Qwen3.6 Max PreviewQwen0.3081556997219652026-04-20evidence
Qwen3.6 PlusQwen0.2784986098239112026-04-02evidence
Qwen3.7 MaxQwen0.4050046339202972026-05-19evidence
Qwen3.7 PlusQwen0.3563484708063022026-06-01evidence
Qwen3.8 2.4T A95BQwen0.4244670991658942026-08-12evidence
Qwen3.8 27B (low)Qwen0.1404077849860982026-08-14evidence
Qwen3.8 27B (medium)Qwen0.1408711770157552026-08-14evidence
Qwen3.8 27B (Non-reasoning)Qwen0.1214087117701582026-08-14evidence
Qwen3.8 27B (xhigh)Qwen0.339202965708992026-08-14evidence
Qwen3.8 MaxQwen0.4304911955514372026-08-03evidence
QwQ 32BQwen0.07352025-03-05evidence
QwQ 32B-PreviewQwen0.03912024-11-27evidence
Reka Flash 3Reka AI0.04442025-03-10evidence
Ring-1TInclusionAI0.1112140871177022025-10-13evidence
Ring-2.6-1TInclusionAI0.2159406858202042026-05-08evidence
Ring-flash-2.0InclusionAI0.09638554216867472025-09-19evidence
Sarvam 105B (high)Sarvam0.1102873030583872026-03-06evidence
Sarvam 30B (high)Sarvam0.07506950880444862026-03-06evidence
Sarvam M (Reasoning)Sarvam0.0310472659870252025-05-23evidence
Seed-OSS-36B-InstructByteDance Seed0.09870250231696022025-08-20evidence
Solar Open 100B (Reasoning)Upstage0.1042632066728452025-12-17evidence
Solar Open2 250BUpstage0.2854494902687672026-08-12evidence
Solar Pro 2 (Non-reasoning)Upstage0.03707136237256722025-07-09evidence
Solar Pro 2 (Preview) (Non-reasoning)Upstage0.03660797034291012025-05-20evidence
Solar Pro 2 (Preview) (Reasoning)Upstage0.06070435588507882025-05-20evidence
Solar Pro 2 (Reasoning)Upstage0.07367933271547732025-07-09evidence
Solar Pro 3Upstage0.1028730305838742026-04-06evidence
Solar Pro 4Upstage0.2924003707136242026-08-06evidence
SonarPerplexity0.04882025-01-21evidence
Sonar ProPerplexity0.06552025-01-21evidence
Step 3.5 FlashStepFun0.2113067655236332026-02-02evidence
Step 3.5 Flash 2603StepFun0.2446709916589432026-04-02evidence
Step 3.7 FlashStepFun0.2140871177015762026-05-29evidence
Step3 VL 10BStepFun0.1084337349397592026-01-20evidence
Tiny Aya GlobalCohere0.05236329935125122026-02-17evidence
Tri-21B-ThinkTrillion Labs0.05931417979610752026-02-10evidence
Tri-21B-think PreviewTrillion Labs0.05792400370713622026-02-10evidence
Trinity Large ThinkingArcee AI0.1584800741427252026-04-01evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.