Changelog

Minor scoring update — no change to Intelligence Index

NVIDIA Nemotron 3 VoiceChat: Leading the Open Weights Frontier of Conversational Dynamics vs. Speech Reasoning

Independent analysis and benchmarks of NVIDIA Nemotron 3 VoiceChat

GPT-5.4 (xhigh): CritPT score updated from 20.0% to 23.4%

What's changed

CritPT score updated for GPT-5.4 (xhigh): We observed higher variance in individual CritPT scores for GPT-5.4 (xhigh) and ran additional evaluation runs to confirm. The CritPT score has been updated from 20.0% to 23.4%.
Intelligence Index unchanged: This does not affect the overall Intelligence Index score for GPT-5.4 (xhigh), as the change is within the Intelligence Index margin of error. All other benchmark scores remain unchanged.

Qwen3.5 397B A17B (Reasoning) on Clarifai

13 Mar 2026

DeepSeek V3.1 (Reasoning) on Eigen AI

DeepSeek V3.2 (Reasoning) on Eigen AI

DeepSeek V3.2 (Non-reasoning) on Eigen AI

Qwen3.5 397B A17B (Reasoning) on GMI

Kimi K2.5 (Non-reasoning) on FriendliAI

Qwen3.5 27B (Reasoning) on GMI

Qwen3.5 35B A3B (Reasoning) on GMI

Qwen3.5 122B A10B (Reasoning) on GMI

Kimi K2.5 (Reasoning) on FriendliAI

Qwen3 Coder Next on Amazon Bedrock

12 Mar 2026

Kimi K2.5 (Reasoning) on Nebius

GLM-5 (Non-reasoning) on Nebius

GLM-5 (Reasoning) on Nebius

11 Mar 2026

NVIDIA Nemotron 3 Super: The new leader in open, efficient intelligence

Independent analysis and benchmarks of NVIDIA Nemotron 3 Super

Grok 4.20 Beta 0309 (Non-reasoning)

Intelligence Index: 30 | AA-Omniscience: -43

Grok 4.20 Beta 0309 (Reasoning)

Intelligence Index: 48 | AA-Omniscience: 13

Qwen3.5 0.8B (Non-reasoning)

Intelligence Index: 10 | AA-Omniscience: -89

NVIDIA Nemotron 3 Super 120B A12B (Reasoning)

Intelligence Index: 36 | AA-Omniscience: -42

LongCat Flash Lite

Intelligence Index: 24 | AA-Omniscience: -70

Grok 4.20 Beta 0309 (Reasoning) on xAI

Grok 4.20 Beta 0309 (Non-reasoning) on xAI

GLM-5 (Non-reasoning) on Eigen AI

GLM-5 (Reasoning) on Eigen AI

Qwen3 235B A22B 2507 Instruct on Eigen AI

DeepSeek V3.1 (Non-reasoning) on Eigen AI

NVIDIA Nemotron 3 Super 120B A12B (Reasoning) on Weights & Biases

NVIDIA Nemotron 3 Super 120B A12B (Reasoning) on Nebius

NVIDIA Nemotron 3 Super 120B A12B (Reasoning) on Baseten

NVIDIA Nemotron 3 Super 120B A12B (Reasoning) on DeepInfra

LongCat Flash Lite on LongCat

10 Mar 2026

Overview of Google I/O Benchmarking Results

Independent benchmarking results covering launches across LLMs, image, video, music and more

Qwen3.5 397B A17B (Reasoning) on Eigen AI

MiniMax-M2.5 on Eigen AI

Llama 4 Maverick on Eigen AI

Kimi K2.5 (Reasoning) on Eigen AI

Qwen3 8B (Non-reasoning) on Eigen AI

Qwen3 30B A3B (Reasoning) on Eigen AI

Qwen3 VL 30B A3B (Reasoning) on Eigen AI

Qwen3 VL 30B A3B Instruct on Eigen AI

Qwen3 8B (Reasoning) on Eigen AI

Qwen3 30B A3B (Non-reasoning) on Eigen AI

Qwen3.5 397B A17B (Non-reasoning) on Eigen AI

Kimi K2.5 (Non-reasoning) on Eigen AI

Kimi K2.5 (Reasoning) on Clarifai

LTX-2.3 Fast

Text to Video ELO: 1127

LTX-2.3 Pro

Text to Video ELO: 1133

LTX-2.3 Fast

Image to Video ELO: 1153

LTX-2.3 Pro

Image to Video ELO: 1165

09 Mar 2026

Qwen3.5 2B (Non-reasoning)

Intelligence Index: 15 | AA-Omniscience: -83

Qwen3.5 4B (Non-reasoning)

Intelligence Index: 23 | AA-Omniscience: -75

Qwen3.5 9B (Non-reasoning)

Intelligence Index: 27 | AA-Omniscience: -71

NVIDIA Nemotron 3 Super 120B A12B (Reasoning) on Lightning AI

DeepSeek V3.2 (Non-reasoning) on FriendliAI

Qwen3.5 9B (Reasoning) on Together.ai

07 Mar 2026

Qwen3 30B A3B 2507 Instruct on Weights & Biases

06 Mar 2026

GPT-5.4 Pro (xhigh)

Kimi K2.5 (Reasoning) on DeepInfra

Qwen2.5 Instruct 72B on SiliconFlow

MiniMax-M2.5 on FriendliAI

MiniMax-M2.1 on FriendliAI

GLM-5 (Reasoning) on FriendliAI

Qwen3 235B A22B 2507 Instruct on FriendliAI

MiniMax-M2.5 on Weights & Biases

Kimi K2.5 (Reasoning) on Weights & Biases

GLM-5 (Reasoning) on Weights & Biases

GPT-5.4 Pro (xhigh) on OpenAI

Step 3.5 Flash on SiliconFlow

Kling Image 3.0

Image Editing ELO: 1203

Kling Image O1

Image Editing ELO: 1191

05 Mar 2026

Qwen3.5 small models: Everything you need to know

Independent analysis and benchmarks of Qwen 3.5 small models

GPT-5.4 (xhigh) on OpenAI

GLM-5 (Reasoning) on Baseten

GLM-5 (Non-reasoning) on Baseten

Llama 3.1 Instruct 8B on Weights & Biases

Qwen3 235B A22B 2507 Instruct on Weights & Biases

Qwen3 Coder 480B A35B Instruct on Weights & Biases

DeepSeek V3.1 (Non-reasoning) on Weights & Biases

gpt-oss-20B (high) on Weights & Biases

gpt-oss-120B (high) on Weights & Biases

Qwen3 235B A22B 2507 (Reasoning) on Weights & Biases

Llama 3.3 Instruct 70B on Weights & Biases

Llama 4 Scout on Weights & Biases

Phi-4 Mini Instruct on Weights & Biases

gpt-oss-120B (low) on Weights & Biases

gpt-oss-20B (low) on Weights & Biases

GLM-4.7 (Non-reasoning) on Amazon Bedrock

GLM-4.7-Flash (Non-reasoning) on Amazon Bedrock

DeepSeek V3.2 (Reasoning) on Amazon Bedrock

DeepSeek V3.2 (Non-reasoning) on Amazon Bedrock

GLM-4.7 (Reasoning) on Amazon Bedrock

GLM-4.7-Flash (Reasoning) on Amazon Bedrock

04 Mar 2026

Qwen3.5 2B (Reasoning)

Intelligence Index: 16 | AA-Omniscience: -47

Qwen3.5 0.8B (Reasoning)

Intelligence Index: 11 | AA-Omniscience: -35

Qwen3.5 4B (Reasoning)

Intelligence Index: 27 | AA-Omniscience: -57

Qwen3.5 9B (Reasoning)

Intelligence Index: 32 | AA-Omniscience: -53

Kimi K2.5 (Reasoning) on Amazon Bedrock

GLM-4.7 (Non-reasoning) on Nebius

Kimi K2.5 (Non-reasoning) on Nebius

Qwen3 235B A22B 2507 (Reasoning) on Nebius

Gemma 3 27B Instruct on Nebius

03 Mar 2026

Gemini 3.1 Flash-Lite Preview

Intelligence Index: 34 | AA-Omniscience: -16

gpt-oss-20B (high) on Nebius

Gemini 3.1 Flash-Lite Preview on Google

gpt-oss-120B (high) on Nebius

02 Mar 2026

Qwen3.5 122B A10B (Non-reasoning)

Intelligence Index: 36 | AA-Omniscience: -54

Qwen3.5 27B (Non-reasoning)

Intelligence Index: 37 | AA-Omniscience: -47

Qwen3.5 35B A3B (Non-reasoning)

Intelligence Index: 31 | AA-Omniscience: -62

Qwen3.5 27B (Non-reasoning) on Alibaba Cloud

Qwen3.5 35B A3B (Non-reasoning) on Alibaba Cloud

Qwen3.5 122B A10B (Non-reasoning) on Alibaba Cloud

Kimi K2.5 (Reasoning) on Lightning AI

P-Video

Text to Video ELO: 1071

Artificial Analysis performance benchmarking methodology

P-Video

Image to Video ELO: 1135

01 Mar 2026

Language model performance benchmarking methodology update (v2.2.0): Expanded prompt set and new default prompt length

What's changed

Refreshed prompt bank: We expanded and diversified prompts across more task types to better reflect real-world usage. This is important because techniques like speculative decoding mean that we see variation in output speeds depending on the type of output.
New default workload: Site-default speed now uses 10k input tokens (was 1k)and we have deprecated performance measurements for 100 token input workloads. 1k, 10k, and 100k remain available via Prompt Options, and the Output Speed by Input Token Count chart is the quickest way to compare workload shapes.

Server location, testing frequency, and median reporting are unchanged.

26 Feb 2026

⚖️ Artificial Analysis Intelligence benchmarking methodology

Updated to Version 4.0.3

Updated grader model for AA-Omniscience to Gemini 3 Flash Preview (Reasoning)

LFM2 24B A2B

Intelligence Index: 10 | AA-Omniscience: -59

GPT-5.3 Codex (xhigh)

Intelligence Index: 54 | AA-Omniscience: 10

LFM2 24B A2B on Together.ai

GPT-5.3 Codex (xhigh) on OpenAI

GLM-5 (Non-reasoning) on Fireworks

Qwen3.5 35B A3B (Reasoning) on Novita

Qwen3.5 27B (Reasoning) on Novita

Qwen3.5 122B A10B (Reasoning) on Novita

Nano Banana 2 (Gemini 3.1 Flash Image Preview)

Text to Image ELO: 1257

Nano Banana 2 (Gemini 3.1 Flash Image Preview)

Image Editing ELO: 1245

25 Feb 2026

Qwen3.5 122B A10B (Reasoning)

Intelligence Index: 42 | AA-Omniscience: -40

Qwen3.5 27B (Reasoning)

Intelligence Index: 42 | AA-Omniscience: -42

Qwen3.5 35B A3B (Reasoning)

Intelligence Index: 37 | AA-Omniscience: -46

Qwen3.5 122B A10B (Reasoning) on Alibaba Cloud

Qwen3.5 27B (Reasoning) on Alibaba Cloud

Qwen3.5 35B A3B (Reasoning) on Alibaba Cloud

PixVerse V5.6

Text to Video ELO: 1220

PixVerse V5.6

Image to Video ELO: 1292

24 Feb 2026

Mercury 2

Intelligence Index: 33 | AA-Omniscience: -52

Kimi K2.5 (Reasoning) on Nebius

Mercury 2 on Inception

Recraft V4

Text to Image ELO: 1118

Recraft V4 Pro

Text to Image ELO: 1129

20 Feb 2026

Tiny Aya Global

Intelligence Index: 5 | AA-Omniscience: -85

Claude Sonnet 4.6 (Non-reasoning, Low Effort)

Intelligence Index: 43 | AA-Omniscience: -2

Gemini 3.1 Pro Preview on Google

19 Feb 2026

Gemini 3.1 Pro Preview

Intelligence Index: 57 | AA-Omniscience: 33

Gemini 3.1 Pro Preview on Google

Kling 3.0 Omni 720p (Standard)

Text to Video ELO: 1218

Kling 3.0 Omni 1080p (Pro)

Text to Video ELO: 1235

Kling 3.0 Omni 720p (Standard)

Image to Video ELO: 1263

Kling 3.0 Omni 1080p (Pro)

Image to Video ELO: 1295

18 Feb 2026

Sonnet 4.6 - Everything you need to know

Independent analysis and benchmarks of Sonnet 4.6

AA-WER v2.0: Speech to Text Accuracy Benchmark

Introducing AA-AgentTalk, a new proprietary evaluation dataset focused on voice agent speech, alongside cleaned ground truth transcripts for VoxPopuli and Earnings22.

Tri-21B-think Preview

Intelligence Index: 20 | AA-Omniscience: -55

Tri-21B-Think

Intelligence Index: 19 | AA-Omniscience: -63

Qwen3.5 397B A17B (Non-reasoning)

Intelligence Index: 40 | AA-Omniscience: -36

Claude Sonnet 4.6 (Non-reasoning, Low Effort) on Anthropic

Qwen3.5 397B A17B (Non-reasoning) on Novita

Qwen3.5 397B A17B (Reasoning) on Parasail

MiniMax-M2.5 on Parasail

MiniMax-M2.5 on Together.ai

GLM-5 (Reasoning) on Together.ai

MiniMax-M2.5 on SiliconFlow

MiniMax-M2.5 on SambaNova

GLM-5 (Reasoning) on Google

Claude Sonnet 4.6 (Non-reasoning, High Effort) on Amazon Bedrock

Claude Sonnet 4.6 (Non-reasoning, High Effort) on Microsoft Azure

Claude Sonnet 4.6 (Non-reasoning, High Effort) on Google

Claude Sonnet 4.6 (Non-reasoning, High Effort) on Anthropic

17 Feb 2026

Claude Sonnet 4.6 - New leader in GDPval-AA

Claude Sonnet 4.6 wins in agentic performance of real-life knowledge work

Qwen3.5-397B-A17B - Everything you need to know

Independent analysis and benchmarks of Qwen3.5-397B-A17B

Claude Sonnet 4.6 (Non-reasoning, High Effort)

Intelligence Index: 44 | AA-Omniscience: -3

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)

Intelligence Index: 52 | AA-Omniscience: 12

Qwen3.5 397B A17B (Reasoning)

Intelligence Index: 45 | AA-Omniscience: -30

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) on Anthropic

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) on Microsoft Azure

Qwen3.5 397B A17B (Reasoning) on Together.ai

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) on Google

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) on Amazon Bedrock

GLM-4.7-Flash (Non-reasoning) on Novita

GLM-5 (Reasoning) on Parasail

GLM-5 (Non-reasoning) on DeepInfra

GLM-5 (Non-reasoning) on SiliconFlow

MiniMax-M2.5 on DeepInfra

GLM-4.7 (Reasoning) on Clarifai

MiniMax-M2.5 on Clarifai

Qwen3.5 397B A17B (Reasoning) on Novita

Qwen3.5 397B A17B (Non-reasoning) on Alibaba Cloud

Qwen3.5 397B A17B (Reasoning) on Alibaba Cloud

16 Feb 2026

Magistral Small 1

AA-Omniscience: -43

GLM-5 (Non-reasoning) on Novita

MiniMax-M2.5 on Fireworks

13 Feb 2026

MiniMax-M2.5: Everything you need to know

Independent analysis and benchmarks of MiniMax-M2.5

GLM-5 (Non-reasoning)

Intelligence Index: 41 | AA-Omniscience: -11

Qwen3 Coder 30B A3B Instruct on Clarifai

MiniMax-M2.5 on Novita

12 Feb 2026

MiniMax-M2.5

Intelligence Index: 42 | AA-Omniscience: -40

GLM-5 (Reasoning) on Fireworks

GLM-5 (Reasoning) on GMI

11 Feb 2026

GLM-5 - Everything you need to know

Independent analysis and benchmarks of GLM-5

GLM-5 (Reasoning)

Intelligence Index: 50 | AA-Omniscience: 2

GLM-5 (Reasoning) on DeepInfra

GLM-5 (Reasoning) on Novita

grok-imagine-image

Text to Image ELO: 1167

grok-imagine-image-pro

Text to Image ELO: 1134

grok-imagine-image

Image Editing ELO: 1225

grok-imagine-image-pro

Image Editing ELO: 1214

Veo 3.1 Fast

Text to Video ELO: 1218

Veo 3.1

Text to Video ELO: 1216

Kling 3.0 1080p (Pro)

Text to Video ELO: 1248

Kling 3.0 720p (Standard)

Text to Video ELO: 1221

Veo 3.1 Fast

Image to Video ELO: 1276

Veo 3.1

Image to Video ELO: 1244

Kling 3.0 1080p (Pro)

Image to Video ELO: 1288

Kling 3.0 720p (Standard)

Image to Video ELO: 1264

10 Feb 2026

MiMo-V2-Flash (Feb 2026)

Intelligence Index: 41 | AA-Omniscience: -18

GLM-5 (Reasoning) on SiliconFlow

08 Feb 2026

MiniMax-M2.5 on MiniMax

06 Feb 2026

Opus 4.6 - Everything you need to know

Independent analysis and benchmarks of Opus 4.6

MiMo-V2-Flash (Feb 2026) on Xiaomi

Z-Image Base

Text to Image ELO: 1042

Qwen Image Plus 2601

Text to Image ELO: 1064

HunyuanImage 3.0 Instruct (Fal)

Text to Image ELO: 1103

Qwen Image Edit Max 2601

Image Editing ELO: 1153

HunyuanImage 3.0 Instruct (Fal)

Image Editing ELO: 1223

05 Feb 2026

Opus 4.6 Takes Lead in Agentic Real-World Knowledge Tasks

We worked with Anthropic to benchmark Claude Opus 4.6 ahead of launch - it reached an Elo of 1606 with adaptive thinking, nearly 150 points ahead of GPT-5.2 (xhigh). This implies a win rate of ~70% when compared head-to-head with OpenAI's December 2025 flagship.

Claude Opus 4.6 (Adaptive Reasoning, Max Effort)

Intelligence Index: 53 | AA-Omniscience: 14

Claude Opus 4.6 (Non-reasoning, High Effort)

Intelligence Index: 46 | AA-Omniscience: 3

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) on Amazon Bedrock

Claude Opus 4.6 (Non-reasoning, High Effort) on Google

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) on Google

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) on Microsoft Azure

Claude Opus 4.6 (Non-reasoning, High Effort) on Microsoft Azure

Qwen3 Coder Next on Parasail

Claude Opus 4.6 (Non-reasoning, High Effort) on Anthropic

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) on Anthropic

Claude Opus 4.6 (Non-reasoning, High Effort) on Amazon Bedrock

Qwen3 Coder Next on Novita

Qwen3 Coder Next on Together.ai

Eigen Image

Text to Image ELO: 1106

Eigen Image

Image Editing ELO: 1164

PixVerse V5.6 (January)

Text to Video ELO: 1134

PixVerse V5.6 (January)

Image to Video ELO: 1253

04 Feb 2026

Qwen3 Coder Next

Intelligence Index: 28 | AA-Omniscience: -61

Step 3.5 Flash

Intelligence Index: 38 | AA-Omniscience: -41

Kimi K2.5 (Non-reasoning) on Baseten

Kimi K2.5 (Reasoning) on Baseten

Kimi K2.5 (Reasoning) on SiliconFlow

Step 3.5 Flash on StepFun

03 Feb 2026

Qwen3 235B A22B 2507 (Reasoning) on Eigen AI

Runway Gen-4.5

Image to Video ELO: 1269

02 Feb 2026

Riverflow 2.0

Text to Image ELO: 1253

Riverflow 2.0

Image Editing ELO: 1284

29 Jan 2026

Qwen3 Max Thinking Benchmarks and Analysis

Independent analysis and benchmarks of Alibaba's Qwen3 Max Thinking

Qwen3 VL 8B Instruct on Together.ai

Kimi K2.5 (Reasoning) on DeepInfra

GLM-4.7-Flash (Reasoning) on DeepInfra

Kimi K2.5 (Non-reasoning) on Kimi

Kimi K2.5 (Non-reasoning) on Together.ai

Kimi K2.5 (Non-reasoning) on GMI

Kimi K2.5 (Non-reasoning) on Novita

Vidu Q3 Pro

Text to Video ELO: 1223

Vidu Q3 Pro

Image to Video ELO: 1287

28 Jan 2026

Kimi K2.5 (Non-reasoning)

Intelligence Index: 37 | AA-Omniscience: -14

Kimi K2.5 (Reasoning) on Parasail

Kimi K2.5 (Reasoning) on Together.ai

Wan2.6 Text to Image

Text to Image ELO: 1134

grok-imagine-video

Text to Video ELO: 1230

Pika 2.5

Text to Video ELO: 1090

grok-imagine-video

Image to Video ELO: 1329

Pika 2.5

Image to Video ELO: 1206

27 Jan 2026

Kimi K2.5 - Everything you need to know

Independent analysis and benchmarks of Moonshot's Kimi K2.5

Qwen3 Max Thinking

Intelligence Index: 40 | AA-Omniscience: -34

Step3 VL 10B

Intelligence Index: 15 | AA-Omniscience: -59

Kimi K2.5 (Reasoning)

Intelligence Index: 47 | AA-Omniscience: -8

K2 Think V2

Intelligence Index: 24 | AA-Omniscience: -34

Qwen3 Max Thinking on Alibaba Cloud

Kimi K2.5 (Reasoning) on Kimi

Kimi K2.5 (Reasoning) on Novita

Kimi K2.5 (Reasoning) on Fireworks

26 Jan 2026

GLM-4.7-Flash (Non-reasoning)

Intelligence Index: 22 | AA-Omniscience: -62

Wan 2.6 Image

Text to Image ELO: 1136

23 Jan 2026

LFM2 1.2B

Intelligence Index: 6 | AA-Omniscience: -72

GLM-4.7 (Reasoning) on Nebius

DeepSeek V3.2 (Reasoning) on Nebius

Kling 2.6 Standard (January)

Text to Video ELO: 1191

Kling 2.6 Pro (January)

Text to Video ELO: 1205

Kling O1 Pro (January)

Text to Video ELO: 1212

Kling O1 Standard (January)

Text to Video ELO: 1189

TeleVideo 2.0

Image to Video ELO: 1269

Kling 2.6 Pro (January)

Image to Video ELO: 1262

Kling 2.6 Standard (January)

Image to Video ELO: 1281

22 Jan 2026

GLM-4.7-Flash (Reasoning)

Intelligence Index: 30 | AA-Omniscience: -59

LFM2.5-1.2B-Thinking

Intelligence Index: 8 | AA-Omniscience: -84

GPT-5.2 Codex (xhigh) on OpenAI

GPT-5.2 Codex (xhigh) on Microsoft Azure

GLM-4.7-Flash (Reasoning) on Novita

21 Jan 2026

GPT-5.2 Codex (xhigh)

Intelligence Index: 49 | AA-Omniscience: -2

Solar Open 100B (Reasoning)

Intelligence Index: 22 | AA-Omniscience: -54

LFM2.5-VL-1.6B

Intelligence Index: 6 | AA-Omniscience: -84

20 Jan 2026

LFM2.5-1.2B-Instruct

Intelligence Index: 8 | AA-Omniscience: -74

19 Jan 2026

Molmo2-8B

Intelligence Index: 7 | AA-Omniscience: -69

Molmo2-8B on Parasail

MiniMax-M2.1 on SiliconFlow

DeepSeek V3.2 (Non-reasoning) on Nebius

16 Jan 2026

Qwen3 Coder 480B A35B Instruct on Eigen AI

Qwen3 Next 80B A3B (Reasoning) on Eigen AI

GLM-4.7 (Non-reasoning) on Google

GLM-4.7 (Reasoning) on Google

GLM-4.7 (Non-reasoning) on Together.ai

GLM-4.7 (Reasoning) on Together.ai

DeepSeek V3.2 (Reasoning) on SambaNova

DeepSeek V3.2 (Non-reasoning) on SambaNova

15 Jan 2026

Ring-1T

Intelligence Index: 23 | AA-Omniscience: -48

Olmo 3.1 32B Instruct

Intelligence Index: 12 | AA-Omniscience: -51

MiniMax-M2.1 on Nebius

Olmo 3.1 32B Instruct on DeepInfra

GLM-Image

Text to Image ELO: 1048

FLUX.2 [klein] 4B

Text to Image ELO: 1069

FLUX.2 [klein] 9B

Text to Image ELO: 1135

FLUX.2 [klein] Base 4B

Text to Image ELO: 983

FLUX.2 [klein] Base 9B

Text to Image ELO: 1076

Qwen Image Max 2512

Text to Image ELO: 1150

GLM-Image

Image Editing ELO: 930

FLUX.2 [klein] 4B

Image Editing ELO: 1107

FLUX.2 [klein] 9B

Image Editing ELO: 1166

FLUX.2 [klein] Base 4B

Image Editing ELO: 1030

FLUX.2 [klein] Base 9B

Image Editing ELO: 1129

Qwen Image on FlyMyAI

FLUX.1 [dev] on FlyMyAI

FLUX.1 [schnell] on FlyMyAI

HunyuanVideo-1.5 (Fal)

Text to Video ELO: 1021

HunyuanVideo-1.5 (Fal)

Image to Video ELO: 1120

14 Jan 2026

Kimi K2 Thinking on Microsoft Azure

DeepSeek V3.2 (Non-reasoning) on Microsoft Azure

Claude Opus 4.5 (Reasoning) on Microsoft Azure

Claude Opus 4.5 (Non-reasoning) on Microsoft Azure

13 Jan 2026

KAT-Coder-Pro V1 on Novita

Z-Image Turbo on fal.ai

Z-Image Turbo on Runware

Z-Image Turbo on Replicate

12 Jan 2026

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) on Nebius

Imagen 4 Ultra

Text to Image ELO: 1174

Imagen 4 Fast

Text to Image ELO: 1065

Imagen 4 Standard

Text to Image ELO: 1095

Qwen Image Edit Plus 2511

Image Editing ELO: 1164

08 Jan 2026

Falcon-H1R-7B

Intelligence Index: 16 | AA-Omniscience: -62

K-EXAONE (Non-reasoning)

Intelligence Index: 23 | AA-Omniscience: -61

New model added to Speech Arena Leaderboard

K-EXAONE (Reasoning)

Intelligence Index: 32 | AA-Omniscience: -58

Step TTS 2

Text to Speech ELO: 1075

07 Jan 2026

Solar Pro 2 (Preview) (Non-reasoning)

Solar Pro 2 (Preview) (Reasoning)

GLM-4.7 (Non-reasoning) on Cerebras

GLM-4.7 (Reasoning) on Cerebras

06 Jan 2026

Ling-flash-2.0

Intelligence Index: 16 | AA-Omniscience: -63

MiniMax-M2.1 on DeepInfra

03 Jan 2026

Mi:dm K 2.5 Pro

Intelligence Index: 23 | AA-Omniscience: -57

02 Jan 2026

GPT-5.2 (medium)

Intelligence Index: 47 | AA-Omniscience: 0

GPT-5.1 (high) on Microsoft Azure

GPT-5.2 (medium) on OpenAI

DeepSeek V3.2 (Reasoning) on Parasail

GLM-4.6V (Reasoning) on Novita

GLM-4.6V (Non-reasoning) on Novita

MiniMax-M2.1 on Novita

01 Jan 2026

Wan 2.6

Text to Video ELO: 1192

Wan 2.6

Image to Video ELO: 1227

31 Dec 2025

Qwen3 Next 80B A3B (Reasoning) on Nebius

30 Dec 2025

Seedance 1.5 pro

Text to Video ELO: 1180

Seedance 1.5 pro

Image to Video ELO: 1254

29 Dec 2025

MiniMax-M2.1 on Fireworks

LongCat Image

Text to Image ELO: 1034

FLUX.2 [dev] Flash

Text to Image ELO: 1138

FLUX.2 [dev] Turbo

Text to Image ELO: 1165

Wan 2.6 Image

Image Editing ELO: 1196

LongCat Image

Image Editing ELO: 1111

FLUX.2 [dev] Flash

Image Editing ELO: 1146

FLUX.2 [dev] Turbo

Image Editing ELO: 1154

Kling 2.6 Pro (December)

Text to Video ELO: 1188

Kling O1 Pro (December)

Text to Video ELO: 1185

Kling 2.6 Pro (December)

Image to Video ELO: 1258

28 Dec 2025

HyperCLOVA X SEED Think (32B)

Intelligence Index: 24 | AA-Omniscience: -53

27 Dec 2025

Mi:dm K 2.5 Pro Preview

AA-Omniscience: -62

GLM-4.7 (Non-reasoning)

Intelligence Index: 34 | AA-Omniscience: -46

GLM-4.7 (Reasoning)

Intelligence Index: 42 | AA-Omniscience: -35

GLM-4.7 (Non-reasoning) on Parasail

GLM-4.7 (Non-reasoning) on Novita

GLM-4.7 (Non-reasoning) on SiliconFlow

GLM-4.7 (Non-reasoning) on DeepInfra

GLM-4.7 (Non-reasoning) on Baseten

GLM-4.7 (Reasoning) on Baseten

GLM-4.7 (Reasoning) on Novita

26 Dec 2025

GLM-4.7 (Reasoning) on Parasail

GLM-4.7 (Reasoning) on DeepInfra

23 Dec 2025

MiniMax-M2.1

Intelligence Index: 39 | AA-Omniscience: -33

Olmo 3.1 32B Think

Intelligence Index: 14 | AA-Omniscience: -44

Olmo 3.1 32B Think on Parasail

19 Dec 2025

MiMo-V2-Flash (Non-reasoning)

Intelligence Index: 30 | AA-Omniscience: -48

MiMo-V2-Flash (Reasoning)

Intelligence Index: 39 | AA-Omniscience: -43

MiMo-V2-Flash (Non-reasoning) on Xiaomi

MiMo-V2-Flash (Reasoning) on Xiaomi

18 Dec 2025

K2-V2 (high)

Intelligence Index: 21 | AA-Omniscience: -57

K2-V2 (low)

Intelligence Index: 14 | AA-Omniscience: -48

K2-V2 (medium)

Intelligence Index: 19 | AA-Omniscience: -50

17 Dec 2025

Gemini 3 Flash - Everything you need to know

Independent analysis and benchmarks of Google's latest Gemini Flash model

Gemini 3 Flash Preview (Non-reasoning)

Intelligence Index: 35 | AA-Omniscience: -4

Gemini 3 Flash Preview (Reasoning)

Intelligence Index: 46 | AA-Omniscience: 12

GLM-4.6V (Reasoning) on DeepInfra

Gemini 3 Flash Preview (Non-reasoning) on Google

Gemini 3 Flash Preview (Reasoning) on Google

16 Dec 2025

GPT-5.2 (Non-reasoning)

Intelligence Index: 34 | AA-Omniscience: -14

15 Dec 2025

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)

Intelligence Index: 24 | AA-Omniscience: -52

NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)

Intelligence Index: 13 | AA-Omniscience: -69

DeepSeek V3.2 (Reasoning) on Fireworks

DeepSeek V3.1 (Reasoning) on Google

DeepSeek V3.2 (Reasoning) on Google

GPT-5.2 (xhigh) on Databricks

DeepSeek V3.2 (Non-reasoning) on Google

Gemini 3 Pro Preview (high) on Databricks

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) on DeepInfra

NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) on DeepInfra

13 Dec 2025

GPT-5.2 (Non-reasoning) on Microsoft Azure

12 Dec 2025

GPT-5.2 (xhigh)

Intelligence Index: 51 | AA-Omniscience: -1

GPT-5.2 (xhigh) on Microsoft Azure

GPT-5.2 (xhigh) on OpenAI

11 Dec 2025

GLM-4.6V (Non-reasoning)

Intelligence Index: 17 | AA-Omniscience: -38

GLM-4.6V (Reasoning)

Intelligence Index: 23 | AA-Omniscience: -26

GLM-4.6V (Non-reasoning) on SiliconFlow

GLM-4.6V (Reasoning) on SiliconFlow

10 Dec 2025

Devstral 2

Intelligence Index: 22 | AA-Omniscience: -46

Devstral Small 2

Intelligence Index: 19 | AA-Omniscience: -57

Nova 2.0 Lite (medium) on Amazon Bedrock

Nova 2.0 Pro Preview (medium) on Amazon Bedrock

Apriel-v1.6-15B-Thinker on Together.ai

Devstral 2 on Mistral

Devstral Small 2 on Mistral

09 Dec 2025

Apriel-v1.6-15B-Thinker

Intelligence Index: 28 | AA-Omniscience: -59

Motif-2-12.7B-Reasoning

Intelligence Index: 19 | AA-Omniscience: -61

08 Dec 2025

KAT-Coder-Pro V1

Intelligence Index: 36 | AA-Omniscience: -37

ERNIE 5.0 Thinking Preview

Intelligence Index: 29 | AA-Omniscience: -44

05 Dec 2025

Ministral 3 3B on Amazon Bedrock

Ministral 3 8B on Amazon Bedrock

Ministral 3 14B on Amazon Bedrock

DeepSeek V3.2 (Non-reasoning) on SiliconFlow

DeepSeek V3.2 (Reasoning) on SiliconFlow

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) on Amazon Bedrock

04 Dec 2025

NVIDIA Nemotron Nano 12B v2 VL (Reasoning)

Intelligence Index: 15 | AA-Omniscience: -64

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

Intelligence Index: 10 | AA-Omniscience: -73

GPT-5 Codex (high) on Microsoft Azure

GPT-5.1 (Non-reasoning) on Microsoft Azure

GPT-5 nano (high) on Databricks

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) on Nebius

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) on DeepInfra

NVIDIA Nemotron Nano 12B v2 VL (Reasoning) on DeepInfra