Benchmark

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

Categories
intelligence-index, knowledge, faithfulness
Openness
unknown
Source
artificial_analysis
Reported scores
489

Reported scores

artificial_analysis

ModelOrganizationReported valueReportedEvidence
A.X-K2SK Telecom0.6700798689330332026-08-12evidence
Agnes 2.5 Pro AlphaSapiens AI0.119268353796041062026-07-24evidence
Apertus 70B InstructSwiss AI Initiative0.17182195397963032025-09-02evidence
Apertus 8B InstructSwiss AI Initiative0.038553772366721972025-09-02evidence
Apriel-v1.5-15B-ThinkerServiceNow0.147217235188509822025-09-30evidence
Apriel-v1.6-15B-ThinkerServiceNow0.08069048574869532025-11-25evidence
Celeris-1Celeris0.072458341134619022026-07-24evidence
Claude 3 HaikuAnthropic0.19546742209631732024-03-04evidence
Claude 3.5 HaikuAnthropic0.58851766513056842024-10-22evidence
Claude 3.7 Sonnet (Non-reasoning)Anthropic0.47191272051996292025-02-24evidence
Claude 3.7 Sonnet (Reasoning)Anthropic0.60027790643816582025-02-24evidence
Claude 4 Sonnet (Non-reasoning)Anthropic0.5903017241379312025-05-22evidence
Claude 4 Sonnet (Reasoning)Anthropic0.70877344255227432025-05-22evidence
Claude 4.5 Haiku (Non-reasoning)Anthropic0.74289053369692252025-10-15evidence
Claude 4.5 Haiku (Reasoning)Anthropic0.72697702785118932025-10-15evidence
Claude 4.5 Sonnet (Non-reasoning)Anthropic0.47288806143821272025-09-29evidence
Claude 4.5 Sonnet (Reasoning)Anthropic0.50918570009930492025-09-29evidence
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic0.363636363636363652026-06-09evidence
Claude Opus 4.5 (Non-reasoning)Anthropic0.237517630465444272025-11-24evidence
Claude Opus 4.5 (Reasoning)Anthropic0.390015600624024962025-11-24evidence
Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic0.37158126375353662026-02-05evidence
Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic0.199200737780510282026-02-05evidence
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic0.57711118356700352026-04-16evidence
Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic0.458986731001206242026-04-16evidence
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic0.60749185667752452026-05-28evidence
Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic0.387920551276854432026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic0.37797956867196372026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic0.39182282793867122026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic0.39324534161490692026-07-24evidence
Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic0.404526748971193452026-07-24evidence
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic0.51550169109357392026-02-17evidence
Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic0.31489015459723362026-02-17evidence
Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic0.39170385928065112026-02-17evidence
Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic0.60633861551292742026-06-30evidence
Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic0.4796072507552872026-06-30evidence
Cogito v2.1 (Reasoning)Deep Cogito0.198376700883265642025-11-18evidence
Command ACohere0.226828782140721552025-03-13evidence
Command A+Cohere0.85842326687397112026-05-20evidence
DeepSeek R1 (Jan '25)DeepSeek0.09714559846485972025-01-20evidence
DeepSeek R1 0528 (May '25)DeepSeek0.165547024952015342025-05-28evidence
DeepSeek R1 0528 Qwen3 8BDeepSeek0.136484434320425232025-05-29evidence
DeepSeek R1 Distill Llama 70BDeepSeek0.189072164948453582025-01-20evidence
DeepSeek V3 (Dec '24)DeepSeek0.099932930918846362024-12-26evidence
DeepSeek V3 0324DeepSeek0.141378550979960332025-03-25evidence
DeepSeek V3.1 (Non-reasoning)DeepSeek0.14288811795316572025-08-21evidence
DeepSeek V3.1 (Reasoning)DeepSeek0.174882629107981252025-08-21evidence
DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek0.12421006755284382025-09-22evidence
DeepSeek V3.1 Terminus (Reasoning)DeepSeek0.25172890733056712025-09-22evidence
DeepSeek V3.2 (Non-reasoning)DeepSeek0.067324561403508732025-12-01evidence
DeepSeek V3.2 (Reasoning)DeepSeek0.17279960218796622025-12-01evidence
DeepSeek V3.2 Exp (Non-reasoning)DeepSeek0.079481641468682532025-09-29evidence
DeepSeek V3.2 Exp (Reasoning)DeepSeek0.19231654014262712025-09-29evidence
DeepSeek V3.2 SpecialeDeepSeek0.108462774504552772025-12-01evidence
DeepSeek V4 Flash (Non-reasoning)DeepSeek0.05450450450450452026-04-24evidence
DeepSeek V4 Flash (Reasoning, High Effort)DeepSeek0.110286591606960082026-04-24evidence
DeepSeek V4 Flash (Reasoning, Max Effort)DeepSeek0.0390398311791083642026-04-24evidence
DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek0.08303047246295782026-07-31evidence
DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek0.085187194791101482026-08-21evidence
DeepSeek V4 Pro (Non-reasoning)DeepSeek0.121504339440694282026-04-24evidence
DeepSeek V4 Pro (Reasoning, High Effort)DeepSeek0.113196814562002262026-04-24evidence
DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek0.059012562080046792026-04-24evidence
DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek0.051735428945645092026-08-13evidence
Devstral 2Mistral0.14733740265207332025-12-09evidence
Devstral MediumMistral0.378056297513868932025-07-10evidence
Devstral Small (Jul '25)Mistral0.23103987490226742025-07-10evidence
Devstral Small (May '25)Mistral0.136489698890649772025-05-21evidence
Devstral Small 2Mistral0.127078384798099742025-12-09evidence
DiffusionGemma 26B A4BGoogle0.091108891108891092026-06-10evidence
Doubao Seed CodeByteDance Seed0.200133126248058572025-11-11evidence
ERNIE 4.5 300B A47BBaidu0.330656784023059542025-06-30evidence
ERNIE 5.0 Thinking PreviewBaidu0.129922945205479422025-11-13evidence
Exaone 4.0 1.2B (Non-reasoning)LG AI Research0.082631578947368392025-07-15evidence
Exaone 4.0 1.2B (Reasoning)LG AI Research0.069946332737030442025-07-15evidence
EXAONE 4.0 32B (Non-reasoning)LG AI Research0.179004288644415422025-07-15evidence
EXAONE 4.0 32B (Reasoning)LG AI Research0.141640866873064992025-07-15evidence
EXAONE 4.5 33BLG AI Research0.200756068444090772026-04-09evidence
Falcon-H1R-7BTII UAE0.112346401404330052026-01-04evidence
G9v3-39A5BAI9Stars0.86961628817541112026-08-20evidence
G9v3-3BAI9Stars0.88338088445078462026-07-23evidence
Gemini 2.0 Flash (Feb '25)Google0.13763066202090592025-02-05evidence
Gemini 2.5 Flash (Non-reasoning)Google0.069945848375451262025-05-20evidence
Gemini 2.5 Flash (Reasoning)Google0.247130317353139732025-05-20evidence
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google0.087927107061503422025-09-25evidence
Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google0.100930232558139492025-09-25evidence
Gemini 2.5 Flash-Lite (Non-reasoning)Google0.122388648009460032025-06-17evidence
Gemini 2.5 Flash-Lite (Reasoning)Google0.22634997969955342025-06-17evidence
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google0.33935158221704522025-09-25evidence
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google0.122129648445437942025-09-08evidence
Gemini 2.5 ProGoogle0.09105824446267432025-06-05evidence
Gemini 3 Flash Preview (Non-reasoning)Google0.075929910851521682025-12-17evidence
Gemini 3 Flash Preview (Reasoning)Google0.07015032211882612025-12-17evidence
Gemini 3 Pro Preview (high)Google0.085499058380414272025-11-18evidence
Gemini 3 Pro Preview (low)Google0.099967752337955522025-11-18evidence
Gemini 3.1 Flash-LiteGoogle0.173378661087866152026-03-03evidence
Gemini 3.1 Pro PreviewGoogle0.49132521225544482026-02-19evidence
Gemini 3.5 Flash (high)Google0.378257887517146732026-05-19evidence
Gemini 3.5 Flash (medium)Google0.38236295539666322026-05-19evidence
Gemini 3.5 Flash (minimal)Google0.25680023398654572026-05-19evidence
Gemini 3.5 Flash-LiteGoogle0.65642722117202272026-07-21evidence
Gemini 3.6 Flash (high)Google0.44370419720186542026-07-21evidence
Gemini 3.7 Flash (high)Google0.35471838866094742026-08-13evidence
Gemini 3.7 Flash (low)Google0.32304379038047382026-08-13evidence
Gemini 3.7 Flash (medium)Google0.3413043478260872026-08-13evidence
Gemma 3 12B InstructGoogle0.0260126347082868532025-03-12evidence
Gemma 3 1B InstructGoogle0.177142857142857162025-03-13evidence
Gemma 3 270MGoogle0.69581301496552882025-08-14evidence
Gemma 3 27B InstructGoogle0.079073329504116392025-03-12evidence
Gemma 3 4B InstructGoogle0.0178829479768786162025-03-12evidence
Gemma 3n E2B InstructGoogle0.06618962432915922025-06-26evidence
Gemma 3n E4B InstructGoogle0.0293425104147799762025-06-26evidence
Gemma 4 12B (Non-reasoning)Google0.265935313031965162026-06-03evidence
Gemma 4 12B (Reasoning)Google0.190203436697610082026-06-03evidence
Gemma 4 26B A4B (Non-reasoning)Google0.079471504634194392026-04-02evidence
Gemma 4 26B A4B (Reasoning)Google0.135558302430984772026-04-02evidence
Gemma 4 31B (Non-reasoning)Google0.18052778888444622026-04-02evidence
Gemma 4 31B (Reasoning)Google0.15006252605252192026-04-02evidence
Gemma 4 E2B (Non-reasoning)Google0.25603338776991472026-04-02evidence
Gemma 4 E2B (Reasoning)Google0.67636558372009992026-04-02evidence
Gemma 4 E4B (Non-reasoning)Google0.462279585529903672026-04-03evidence
Gemma 4 E4B (Reasoning)Google0.69061075660893352026-04-03evidence
GLM 5V Turbo (Reasoning)Z.ai0.31202830188679242026-04-01evidence
GLM-4.5 (Reasoning)Z.ai0.298530721282279642025-07-28evidence
GLM-4.5-AirZ.ai0.071243781094527342025-07-28evidence
GLM-4.5V (Non-reasoning)Z.ai0.097615651110658292025-08-11evidence
GLM-4.5V (Reasoning)Z.ai0.152210526315789462025-08-11evidence
GLM-4.6 (Non-reasoning)Z.ai0.324284199363732762025-09-30evidence
GLM-4.6 (Reasoning)Z.ai0.059493959425575562025-09-30evidence
GLM-4.6V (Non-reasoning)Z.ai0.333736884584342252025-12-08evidence
GLM-4.6V (Reasoning)Z.ai0.48538476834360712025-12-08evidence
GLM-4.7 (Non-reasoning)Z.ai0.070929725010912212025-12-22evidence
GLM-4.7 (Reasoning)Z.ai0.069794859702900222025-12-22evidence
GLM-4.7-Flash (Non-reasoning)Z.ai0.0571209507379719652026-01-19evidence
GLM-4.7-Flash (Reasoning)Z.ai0.060624130391572222026-01-19evidence
GLM-5 (Non-reasoning)Z.ai0.54710851202079282026-02-11evidence
GLM-5 (Reasoning)Z.ai0.64676616915422882026-02-11evidence
GLM-5-TurboZ.ai0.37363149312834842026-03-15evidence
GLM-5.1 (Non-reasoning)Z.ai0.363555357540654932026-04-07evidence
GLM-5.1 (Reasoning)Z.ai0.7005242463958062026-04-07evidence
GLM-5.2 (max)Z.ai0.73700440528634362026-06-16evidence
GLM-5.2 (Non-reasoning)Z.ai0.66318210328245872026-06-16evidence
GLM-5.3 (max)Z.ai0.70445956160241872026-08-18evidence
GPT-4OpenAI0.305907172995780632023-03-14evidence
GPT-4.1OpenAI0.066912782648823282025-04-14evidence
GPT-4.1 miniOpenAI0.072788119640242682025-04-14evidence
GPT-4.1 nanoOpenAI0.174357984166827572025-04-14evidence
GPT-4o (Aug '24)OpenAI0.415902140672782862024-08-06evidence
GPT-4o (Nov '24)OpenAI0.62084026622296172024-11-20evidence
GPT-5 (high)OpenAI0.178162524434515522025-08-07evidence
GPT-5 (low)OpenAI0.209803393482359322025-08-07evidence
GPT-5 (medium)OpenAI0.167997796750206542025-08-07evidence
GPT-5 (minimal)OpenAI0.098531501657981972025-08-07evidence
GPT-5 Codex (high)OpenAI0.245785753126699262025-09-23evidence
GPT-5 mini (high)OpenAI0.43647267880941812025-08-07evidence
GPT-5 mini (medium)OpenAI0.55675909878682852025-08-07evidence
GPT-5 mini (minimal)OpenAI0.106597938144329942025-08-07evidence
GPT-5 nano (high)OpenAI0.40988671472708552025-08-07evidence
GPT-5 nano (medium)OpenAI0.47381193124368052025-08-07evidence
GPT-5 nano (minimal)OpenAI0.112368168744007682025-08-07evidence
GPT-5.1 (high)OpenAI0.48113459994648112025-11-13evidence
GPT-5.1 (Non-reasoning)OpenAI0.09380907372400762025-11-13evidence
GPT-5.1 Codex (high)OpenAI0.22795341098169722025-11-13evidence
GPT-5.1 Codex mini (high)OpenAI0.484157986111111162025-11-13evidence
GPT-5.2 (medium)OpenAI0.3840129624628682025-12-11evidence
GPT-5.2 (Non-reasoning)OpenAI0.376717281272595852025-12-11evidence
GPT-5.2 (xhigh)OpenAI0.188266985932355582025-12-11evidence
GPT-5.2 Codex (xhigh)OpenAI0.266402714932126752025-12-11evidence
GPT-5.3 Codex (xhigh)OpenAI0.10824195259992932026-02-05evidence
GPT-5.4 (low)OpenAI0.17417705337168422026-03-05evidence
GPT-5.4 (Non-reasoning)OpenAI0.15278147458078252026-03-05evidence
GPT-5.4 (xhigh)OpenAI0.083079009833841982026-03-05evidence
GPT-5.4 mini (medium)OpenAI0.101521511017838352026-03-17evidence
GPT-5.4 mini (Non-Reasoning)OpenAI0.044160502129567352026-03-17evidence
GPT-5.4 mini (xhigh)OpenAI0.09834754797441372026-03-17evidence
GPT-5.4 nano (medium)OpenAI0.48868971404182672026-03-17evidence
GPT-5.4 nano (Non-Reasoning)OpenAI0.38476600744076762026-03-17evidence
GPT-5.4 nano (xhigh)OpenAI0.258295964125560552026-03-17evidence
GPT-5.5 (high)OpenAI0.10899922420480992026-04-23evidence
GPT-5.5 (low)OpenAI0.119601328903654492026-04-23evidence
GPT-5.5 (medium)OpenAI0.104166666666666632026-04-23evidence
GPT-5.5 (Non-reasoning)OpenAI0.073267933782955262026-04-23evidence
GPT-5.5 (xhigh)OpenAI0.1097899326198972026-04-23evidence
GPT-5.5 Instant (June 2026)OpenAI0.26746626686656672026-06-25evidence
GPT-5.5 Instant (May 2026)OpenAI0.34697674418604652026-05-05evidence
GPT-5.6 Luna (high)OpenAI0.07615230460921842026-07-09evidence
GPT-5.6 Luna (low)OpenAI0.101297267457907792026-07-09evidence
GPT-5.6 Luna (max)OpenAI0.074214202561117552026-07-09evidence
GPT-5.6 Luna (medium)OpenAI0.091343451377178142026-07-09evidence
GPT-5.6 Luna (Non-reasoning)OpenAI0.252026-07-09evidence
GPT-5.6 Luna (xhigh)OpenAI0.075296843324645262026-07-09evidence
GPT-5.6 Sol (high)OpenAI0.08803521408563432026-07-09evidence
GPT-5.6 Sol (low)OpenAI0.105836575875486342026-07-09evidence
GPT-5.6 Sol (max)OpenAI0.077996715927750372026-07-09evidence
GPT-5.6 Sol (medium)OpenAI0.092307692307692312026-07-09evidence
GPT-5.6 Sol (Non-reasoning)OpenAI0.07212475633528272026-07-09evidence
GPT-5.6 Sol (xhigh)OpenAI0.081343585592877352026-07-09evidence
GPT-5.6 Terra (high)OpenAI0.102109446652399852026-07-09evidence
GPT-5.6 Terra (low)OpenAI0.1013033175355452026-07-09evidence
GPT-5.6 Terra (max)OpenAI0.121240601503759372026-07-09evidence
GPT-5.6 Terra (medium)OpenAI0.102286401925391072026-07-09evidence
GPT-5.6 Terra (Non-reasoning)OpenAI0.049854919546293842026-07-09evidence
GPT-5.6 Terra (xhigh)OpenAI0.109819516671765042026-07-09evidence
gpt-oss-120b (high)OpenAI0.091838909013424292025-08-05evidence
gpt-oss-120b (low)OpenAI0.086034912718204512025-08-05evidence
gpt-oss-20b (high)OpenAI0.058928571428571442025-08-05evidence
gpt-oss-20b (low)OpenAI0.132875368007850822025-08-05evidence
Granite 3.3 8B (Non-reasoning)IBM0.037815901125253682025-04-16evidence
Granite 4.0 1BIBM0.06481974782454272025-10-28evidence
Granite 4.0 350MIBM0.23868954758190332025-10-28evidence
Granite 4.0 H 1BIBM0.182841068917018232025-10-28evidence
Granite 4.0 H 350MIBM0.119369369369369372025-10-28evidence
Granite 4.0 H SmallIBM0.121524402099941642025-09-22evidence
Granite 4.0 MicroIBM0.042271641242418672025-09-22evidence
Granite 4.1 30BIBM0.050164632965330252026-04-29evidence
Granite 4.1 3BIBM0.049219467401285582026-04-29evidence
Granite 4.1 8BIBM0.128658304827061982026-04-29evidence
Grok 3xAI0.126816690107829372025-02-19evidence
Grok 3 mini Reasoning (high)xAI0.74180247398389952025-02-19evidence
Grok 4xAI0.355082609913189632025-07-10evidence
Grok 4 Fast (Non-reasoning)xAI0.123531794248683662025-09-19evidence
Grok 4 Fast (Reasoning)xAI0.31728901359810062025-09-19evidence
Grok 4.1 Fast (Non-reasoning)xAI0.177003624647603662025-11-19evidence
Grok 4.1 Fast (Reasoning)xAI0.265850945494994442025-11-19evidence
Grok 4.20 0309 (Non-reasoning)xAI0.071895424836601272026-03-10evidence
Grok 4.20 0309 (Reasoning)xAI0.77595500351535042026-03-10evidence
Grok 4.20 0309 v2 (Non-reasoning)xAI0.0272789270288702482026-04-07evidence
Grok 4.20 0309 v2 (Reasoning)xAI0.82555734313950812026-04-07evidence
Grok 4.3 (high)xAI0.74239713774597492026-04-30evidence
Grok 4.3 (low)xAI0.82609684019095252026-04-30evidence
Grok 4.3 (medium)xAI0.83060364997660272026-04-30evidence
Grok 4.3 (Non-reasoning)xAI0.25540747214332542026-04-30evidence
Grok 4.5 (high)xAI0.458548331613347072026-07-08evidence
Grok 4.6 (high)xAI0.65711526078557642026-08-12evidence
Grok 4.6 (low)xAI0.69350573023802522026-08-12evidence
Grok 4.6 (medium)xAI0.76004592422502872026-08-12evidence
Grok 4.6 (xhigh)xAI0.75994152046783632026-08-12evidence
Grok Build 0.1 0616xAI0.068752148504640732026-06-16evidence
Grok Code Fast 1xAI0.207107041639415732025-08-28evidence
Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research0.197656073923822452025-08-27evidence
Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research0.0545887961859355952025-08-27evidence
Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research0.1731557377049182025-08-27evidence
Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research0.046217571397427552025-08-27evidence
Hy3Tencent0.25869671729544342026-07-06evidence
Hy3-preview (Non-reasoning)Tencent0.243219787372532032026-04-23evidence
Hy3-preview (Reasoning)Tencent0.126040703052728962026-04-23evidence
HyperCLOVA X SEED Think (32B)Naver0.188140267927501942025-12-26evidence
HyperNova 60B 2605Multiverse Computing0.176724985256536242026-05-26evidence
Inkling (xhigh)Thinking Machines0.323353293413173652026-07-15evidence
Inkling SmallThinking Machines0.370074812967581052026-07-30evidence
INTELLECT-3Prime Intellect0.124145077720207222025-11-27evidence
Jamba 1.7 LargeAI21 Labs0.049317943336831042025-07-07evidence
Jamba 1.7 MiniAI21 Labs0.049023290758827942025-07-07evidence
Jamba Reasoning 3BAI21 Labs0.30957484816005722025-10-08evidence
JT-35B-FlashChina Mobile0.36161793800835352026-05-14evidence
JT-4.1 Flash 236B A21BChina Mobile0.5632841734536652026-07-09evidence
JT-MINIChina Mobile0.087040994754225782026-04-15evidence
K-EXAONE (Non-reasoning)LG AI Research0.084539664157498582025-12-31evidence
K-EXAONE (Reasoning)LG AI Research0.111576011157601072025-12-31evidence
K-EXAONE 2.0 0803LG AI Research0.77400805060379532026-08-12evidence
K2 Think V2MBZUAI Institute of Foundation Models0.288269973571864242025-12-15evidence
K2-V2 (high)MBZUAI Institute of Foundation Models0.090220949263502442025-12-05evidence
K2-V2 (low)MBZUAI Institute of Foundation Models0.239641434262948262025-12-05evidence
K2-V2 (medium)MBZUAI Institute of Foundation Models0.184402924451665282025-12-05evidence
KAT Coder Pro V2KwaiKAT0.440246232222458072026-03-27evidence
KAT-Coder-Pro V1KwaiKAT0.32644795463750512025-11-11evidence
Kimi K2Moonshot AI0.233769213122275722025-07-11evidence
Kimi K2 0905Moonshot AI0.302793296089385432025-09-05evidence
Kimi K2 ThinkingMoonshot AI0.2437319189971072025-11-06evidence
Kimi K2.5 (Non-reasoning)Moonshot AI0.50043936731107212026-01-27evidence
Kimi K2.5 (Reasoning)Moonshot AI0.34319526627218932026-01-27evidence
Kimi K2.6Moonshot AI0.59495548961424332026-04-20evidence
Kimi K2.6 (Non-reasoning)Moonshot AI0.55768383971818582026-04-20evidence
Kimi K2.7 CodeMoonshot AI0.17622724765581912026-06-12evidence
Kimi K3 (low)Moonshot AI0.228878648233486912026-07-16evidence
Kimi K3 (max)Moonshot AI0.468044515103338672026-07-16evidence
LFM2 1.2BLiquid AI0.186185133239831662025-07-10evidence
LFM2 2.6BLiquid AI0.371101321585903032025-09-23evidence
LFM2 24B A2BLiquid AI0.30969351389878832026-02-25evidence
LFM2 8B A1BLiquid AI0.076992918104231032025-10-07evidence
LFM2.5-1.2B-InstructLiquid AI0.149130668578598362026-01-05evidence
LFM2.5-1.2B-ThinkingLiquid AI0.0563456345634563152026-01-20evidence
LFM2.5-2.6BLiquid AI0.8400418191322532026-08-04evidence
LFM2.5-8B-A1BLiquid AI0.53088235294117652026-05-28evidence
LFM2.5-VL-1.6BLiquid AI0.0426322306739783662026-01-05evidence
Ling 2.6 FlashInclusionAI0.0327610025656206452026-04-21evidence
Ling 3.0 FlashInclusionAI0.5594704684317722026-08-04evidence
Ling 3.0 TinyInclusionAI0.69539078156312622026-08-06evidence
Ling-1TInclusionAI0.0527196652719664762025-10-08evidence
Ling-2.6-1TInclusionAI0.071702944942381612026-04-23evidence
Ling-flash-2.0InclusionAI0.105201863354037252025-09-17evidence
Ling-mini-2.0InclusionAI0.041338942747393472025-09-09evidence
Llama 3 Instruct 70BMeta0.124569576665991442024-04-18evidence
Llama 3 Instruct 8BMeta0.10208255857195982024-04-18evidence
Llama 3.1 Instruct 405BMeta0.47559123454111522024-07-23evidence
Llama 3.1 Instruct 70BMeta0.21793275217932752024-07-23evidence
Llama 3.1 Instruct 8BMeta0.56958105646630242024-07-23evidence
Llama 3.1 Nemotron Instruct 70BNVIDIA0.28739359546007292024-10-15evidence
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA0.187656380316930772025-04-07evidence
Llama 3.2 Instruct 11B (Vision)Meta0.181208053691275132024-09-25evidence
Llama 3.2 Instruct 1BMeta0.33655913978494622024-09-25evidence
Llama 3.3 Instruct 70BMeta0.097881965864692562024-12-06evidence
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA0.26309707241910632025-03-18evidence
Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIA0.090470446320868542025-03-18evidence
Llama 4 MaverickMeta0.111185086551265022025-04-05evidence
Llama 4 ScoutMeta0.206483300589391022025-04-05evidence
Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA0.327421815408085462025-07-25evidence
Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA0.218108651911468782025-07-25evidence
LongCat 2.0LongCat0.246091899573661732026-06-29evidence
LongCat Flash LiteLongCat0.038446751249519462026-01-28evidence
Magistral Medium 1Mistral0.40669456066945612025-06-10evidence
Magistral Medium 1.2Mistral0.400168314748579832025-09-18evidence
Magistral Small 1Mistral0.23866639560886362025-06-10evidence
Magistral Small 1.2Mistral0.092432120161756172025-09-17evidence
Mercury 2Inception0.087756396701205272026-02-20evidence
Mi:dm K 2.5 ProKorea Telecom0.061497868885731722025-12-11evidence
Mi:dm K 2.5 Pro PreviewKorea Telecom0.053040877367896312025-12-11evidence
MiMo-V2-Flash (Feb 2026)Xiaomi0.51589445252441312025-12-16evidence
MiMo-V2-Flash (Non-reasoning)Xiaomi0.24041880679573292025-12-16evidence
MiMo-V2-Flash (Reasoning)Xiaomi0.06629711751662972025-12-16evidence
MiMo-V2-OmniXiaomi0.51135894258570842026-03-19evidence
MiMo-V2-Omni-0327Xiaomi0.59947272358547952026-03-27evidence
MiMo-V2-ProXiaomi0.7002044060867592026-03-18evidence
MiMo-V2.5Xiaomi0.68068068068068072026-04-22evidence
MiMo-V2.5-ProXiaomi0.75295381310418912026-04-22evidence
MiMo-V2.5-Pro (Non-reasoning)Xiaomi0.107576104371709722026-04-22evidence
MiniCPM-V 4.6 1.3BOpenBMB0.039313795568263092026-05-11evidence
MiniCPM5-1B (Non-reasoning)OpenBMB0.9909909909909912026-05-25evidence
MiniCPM5-1B (Reasoning)OpenBMB0.83004423273222182026-05-25evidence
MiniMax M1 80kMiniMax0.084694754944110052025-06-17evidence
MiniMax-M2MiniMax0.09096830221450282025-10-26evidence
MiniMax-M2.1MiniMax0.314587737843551762025-12-23evidence
MiniMax-M2.5MiniMax0.118762700383833852026-02-12evidence
MiniMax-M2.7MiniMax0.64435336976320582026-03-18evidence
MiniMax-M3MiniMax0.81572629051620652026-06-01evidence
Ministral 3 14BMistral0.0748312439729992025-12-02evidence
Ministral 3 3BMistral0.198132210217908832025-12-02evidence
Ministral 3 8BMistral0.058417927600076582025-12-02evidence
Mistral 7B InstructMistral0.187396959150027522023-09-27evidence
Mistral Large 2 (Nov '24)Mistral0.32321131447587362024-11-18evidence
Mistral Large 3Mistral0.140350877192982452025-12-02evidence
Mistral Medium 3Mistral0.391020408163265332025-05-07evidence
Mistral Medium 3.1Mistral0.154153522607781262025-08-12evidence
Mistral Medium 3.5Mistral0.1840707964601772026-04-29evidence
Mistral Small 3.1Mistral0.224381625441696152025-03-17evidence
Mistral Small 3.2Mistral0.231898238747553782025-06-20evidence
Mistral Small 4 (Non-reasoning)Mistral0.219936076707950462026-03-16evidence
Mistral Small 4 (Reasoning)Mistral0.33496488614598852026-03-16evidence
Molmo2-8BAllen Institute for AI0.089026915113871622025-12-11evidence
Motif 3Motif Technologies0.71568160152526222026-08-12evidence
Motif 3 (Beta)Motif Technologies0.50896813673770842026-07-14evidence
Motif-2-12.7B-ReasoningMotif Technologies0.094106090373280992025-12-04evidence
Muse Glimmer (high)Meta0.180552385300159822026-08-10evidence
Muse SparkMeta0.158068783068783032026-04-08evidence
Muse Spark 1.1 (xhigh)Meta0.5001737921445952026-07-09evidence
Muse Spark 1.2 (xhigh)Meta0.66707354287458042026-08-05evidence
Nanbeige4.1-3BNanbeige0.419008875739644962026-02-11evidence
Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA0.14347484276729562026-04-29evidence
Nemotron 3 Super 120B A12B (Reasoning)NVIDIA0.130367760405197042026-03-11evidence
Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA0.70312163616792242026-06-04evidence
Nemotron 3.5 LightningNVIDIA0.62426957537982092026-08-11evidence
Nemotron Cascade 2 30B A3BNVIDIA0.146440954517642272026-03-19evidence
Nex-N2-ProNex AGI0.0508821273331628852026-06-02evidence
North Mini CodeCohere0.167899712289354742026-06-09evidence
Nova 2.0 Lite (high)Amazon0.085837793330588722025-10-29evidence
Nova 2.0 Lite (low)Amazon0.128718882817243422025-10-29evidence
Nova 2.0 Lite (medium)Amazon0.081873727087576342025-10-29evidence
Nova 2.0 Lite (Non-reasoning)Amazon0.14130645473929062025-10-29evidence
Nova 2.0 Omni (low)Amazon0.152979725578537782025-11-26evidence
Nova 2.0 Omni (medium)Amazon0.067031186715269312025-11-26evidence
Nova 2.0 Omni (Non-reasoning)Amazon0.107184018440261262025-11-26evidence
Nova 2.0 Pro Preview (low)Amazon0.30478488138319262025-11-27evidence
Nova 2.0 Pro Preview (medium)Amazon0.089918838103374642025-11-27evidence
Nova 2.0 Pro Preview (Non-reasoning)Amazon0.200601805416248722025-11-27evidence
Nova LiteAmazon0.423872875092387252024-12-03evidence
Nova MicroAmazon0.349203408669877762024-12-03evidence
Nova PremierAmazon0.28084656084656092025-04-30evidence
Nova ProAmazon0.222935044105854072024-12-03evidence
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA0.089883414817600652025-12-15evidence
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA0.167271261588069382025-12-15evidence
NVIDIA Nemotron 3 Nano 4BNVIDIA0.11426465167711022026-03-16evidence
NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA0.044654683065279092025-10-28evidence
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA0.076517665430411832025-10-28evidence
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA0.25632969876178162025-08-18evidence
NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA0.395124716553287962025-08-18evidence
o1OpenAI0.30371690427698572024-12-05evidence
o3OpenAI0.119099294628323362025-04-16evidence
o3-mini (high)OpenAI0.18924640135478412025-01-31evidence
o4-mini (high)OpenAI0.195478723404255322025-04-16evidence
Olmo 3 32B ThinkAllen Institute for AI0.143304620203602242025-11-20evidence
Olmo 3 7B InstructAllen Institute for AI0.087832973362131032025-11-20evidence
Olmo 3 7B ThinkAllen Institute for AI0.06785379568884722025-11-20evidence
Olmo 3.1 32B InstructAllen Institute for AI0.26757627439833242026-01-13evidence
Olmo 3.1 32B ThinkAllen Institute for AI0.33713512468586892025-12-12evidence
Phi-4Microsoft0.188130333591931762024-12-12evidence
Phi-4 Mini InstructMicrosoft0.22007366482504612024-02-26evidence
Qwen2.5 Instruct 72BQwen0.14539579967689822024-09-19evidence
Qwen3 0.6B (Non-reasoning)Qwen0.061659388646288172025-04-28evidence
Qwen3 0.6B (Reasoning)Qwen0.09056939501779362025-04-28evidence
Qwen3 1.7B (Non-reasoning)Qwen0.0223813786929274452025-04-28evidence
Qwen3 1.7B (Reasoning)Qwen0.051938551572787152025-04-28evidence
Qwen3 14B (Non-reasoning)Qwen0.076923076923076872025-04-28evidence
Qwen3 14B (Reasoning)Qwen0.241786346645681662025-04-28evidence
Qwen3 235B A22B (Non-reasoning)Qwen0.128735161686451052025-04-28evidence
Qwen3 235B A22B (Reasoning)Qwen0.223813420621931212025-04-28evidence
Qwen3 235B A22B 2507 (Reasoning)Qwen0.101705895055063672025-07-25evidence
Qwen3 235B A22B 2507 InstructQwen0.22826086956521742025-07-21evidence
Qwen3 30B A3B (Non-reasoning)Qwen0.114539946989776652025-04-28evidence
Qwen3 30B A3B (Reasoning)Qwen0.19268244183734342025-04-28evidence
Qwen3 30B A3B 2507 (Reasoning)Qwen0.134607726005575472025-07-30evidence
Qwen3 30B A3B 2507 InstructQwen0.046060368482947812025-07-29evidence
Qwen3 32B (Reasoning)Qwen0.17884537747274932025-04-28evidence
Qwen3 4B 2507 (Reasoning)Qwen0.167682926829268332025-08-06evidence
Qwen3 4B 2507 InstructQwen0.182566404788626962025-08-06evidence
Qwen3 8B (Non-reasoning)Qwen0.04367385192127462025-04-28evidence
Qwen3 8B (Reasoning)Qwen0.092628328830567372025-04-28evidence
Qwen3 Coder 30B A3B InstructQwen0.196969696969697022025-07-31evidence
Qwen3 Coder 480B A35B InstructQwen0.55070171970745212025-07-22evidence
Qwen3 Coder NextQwen0.063208109719737672026-02-03evidence
Qwen3 MaxQwen0.10077177508269022025-09-23evidence
Qwen3 Max (Preview)Qwen0.116314863676341252025-09-05evidence
Qwen3 Max ThinkingQwen0.053431488957492282026-01-26evidence
Qwen3 Max Thinking (Preview)Qwen0.085111264051387912025-11-03evidence
Qwen3 Next 80B A3B (Reasoning)Qwen0.129888843145327342025-09-11evidence
Qwen3 Next 80B A3B InstructQwen0.072507552870090592025-09-11evidence
Qwen3 Omni 30B A3B (Reasoning)Qwen0.110070257611241232025-09-22evidence
Qwen3 Omni 30B A3B InstructQwen0.0241245136186770682025-09-22evidence
Qwen3 VL 235B A22B (Reasoning)Qwen0.14845230574857862025-09-23evidence
Qwen3 VL 235B A22B InstructQwen0.084483479715600222025-09-23evidence
Qwen3 VL 30B A3B (Reasoning)Qwen0.10232744783306582025-10-03evidence
Qwen3 VL 30B A3B InstructQwen0.063401145565870072025-10-03evidence
Qwen3 VL 32B (Reasoning)Qwen0.163187474909674782025-10-21evidence
Qwen3 VL 32B InstructQwen0.081800391389432512025-10-21evidence
Qwen3 VL 4B (Reasoning)Qwen0.08183026390734762025-10-14evidence
Qwen3 VL 4B InstructQwen0.0258040388930441682025-10-14evidence
Qwen3 VL 8B (Reasoning)Qwen0.085881860075408432025-10-14evidence
Qwen3 VL 8B InstructQwen0.089918256130790192025-10-14evidence
Qwen3.5 0.8B (Non-reasoning)Qwen0.0152604806174355682026-03-02evidence
Qwen3.5 0.8B (Reasoning)Qwen0.386529759832927262026-03-02evidence
Qwen3.5 122B A10B (Non-reasoning)Qwen0.08407170822171852026-02-24evidence
Qwen3.5 122B A10B (Reasoning)Qwen0.1291317761128252026-02-24evidence
Qwen3.5 27B (Non-reasoning)Qwen0.248665744218224962026-02-24evidence
Qwen3.5 27B (Reasoning)Qwen0.18508403361344542026-02-24evidence
Qwen3.5 2B (Non-reasoning)Qwen0.022674104732769432026-03-02evidence
Qwen3.5 2B (Reasoning)Qwen0.25761421319796952026-03-02evidence
Qwen3.5 35B A3B (Non-reasoning)Qwen0.065849317777338382026-02-24evidence
Qwen3.5 35B A3B (Reasoning)Qwen0.145659432387312162026-02-24evidence
Qwen3.5 397B A17B (Non-reasoning)Qwen0.17310664605873262026-02-16evidence
Qwen3.5 397B A17B (Reasoning)Qwen0.110549132947976862026-02-16evidence
Qwen3.5 4B (Non-reasoning)Qwen0.0319128965646705562026-03-02evidence
Qwen3.5 4B (Reasoning)Qwen0.134498331042607562026-03-02evidence
Qwen3.5 9B (Non-reasoning)Qwen0.01550087192404572026-03-02evidence
Qwen3.5 9B (Reasoning)Qwen0.16390827517447662026-03-02evidence
Qwen3.5 Omni FlashQwen0.066002733841046712026-03-30evidence
Qwen3.5 Omni PlusQwen0.62999596285829632026-03-30evidence
Qwen3.6 27B (Non-reasoning)Qwen0.160496297778667252026-04-22evidence
Qwen3.6 27B (Reasoning)Qwen0.50725538971807632026-04-22evidence
Qwen3.6 35B A3B (Non-reasoning)Qwen0.078647188312987782026-04-16evidence
Qwen3.6 35B A3B (Reasoning)Qwen0.49497023198521862026-04-16evidence
Qwen3.6 Max PreviewQwen0.53809012875536482026-04-20evidence
Qwen3.6 PlusQwen0.65361104822277572026-04-02evidence
Qwen3.7 MaxQwen0.74370764762826712026-05-19evidence
Qwen3.7 PlusQwen0.72338137233813732026-06-01evidence
Qwen3.8 2.4T A95BQwen0.60824242424242432026-08-12evidence
Qwen3.8 27B (low)Qwen0.47144006436041832026-08-14evidence
Qwen3.8 27B (medium)Qwen0.332516840171463552026-08-14evidence
Qwen3.8 27B (Non-reasoning)Qwen0.81914311759343662026-08-14evidence
Qwen3.8 27B (xhigh)Qwen0.69713721618953612026-08-14evidence
Qwen3.8 MaxQwen0.58253851797505512026-08-03evidence
Reka Flash 3Reka AI0.087258687258687242025-03-10evidence
Ring-1TInclusionAI0.102472757753562462025-10-13evidence
Ring-2.6-1TInclusionAI0.146702557200538372026-05-08evidence
Ring-flash-2.0InclusionAI0.099261034551627712025-09-19evidence
Sarvam 105B (high)Sarvam0.065736245954692522026-03-06evidence
Sarvam 30B (high)Sarvam0.037192447072286862026-03-06evidence
Sarvam M (Reasoning)Sarvam0.093590247738891022025-05-23evidence
Seed-OSS-36B-InstructByteDance Seed0.136104319478402622025-08-20evidence
Solar Open 100B (Reasoning)Upstage0.1035188216039282025-12-17evidence
Solar Open2 250BUpstage0.74620123203285422026-08-12evidence
Solar Pro 2 (Non-reasoning)Upstage0.069684335914234622025-07-09evidence
Solar Pro 2 (Reasoning)Upstage0.0526754112013324562025-07-09evidence
Solar Pro 3Upstage0.117791411042944772026-04-06evidence
Solar Pro 4Upstage0.75596217105263162026-08-06evidence
Step 3.5 FlashStepFun0.143355880427667452026-02-02evidence
Step 3.5 Flash 2603StepFun0.086889232810081762026-04-02evidence
Step 3.7 FlashStepFun0.15023579609252192026-05-29evidence
Step3 VL 10BStepFun0.166794773251345172026-01-20evidence
Tiny Aya GlobalCohere0.037970191625266162026-02-17evidence
Tri-21B-ThinkTrillion Labs0.14581749049429662026-02-10evidence
Tri-21B-think PreviewTrillion Labs0.19466073414905452026-02-10evidence
Trinity Large ThinkingArcee AI0.141075268817204342026-04-01evidence

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.