Scores checked 30 October 2026
Every AI model, side by side
71 text models, 8 image models and 7 video models, each with its own page: what it is, how it scores, where to get an API key and which ones are free. Then use them all at once on one screen.
The scoreboard
Every score comes from one independent source, the Artificial Analysis Intelligence Index, at the best setting it lists. One source keeps the numbers fair to compare. Models it has not scored say so; we never fill in a guess.
71 of 71 models
| Maker | Free API | ||||
|---|---|---|---|---|---|
| Gemini 4 CarbonReasoning | 59 | n/a | n/a | Paid | |
| Claude Opus 5.5Reasoning | Anthropic | 58 | 1M | 97/s | Paid |
| GPT-6 AstraReasoning | OpenAI | 57 | 1M | 76/s | Paid |
| GPT-6.1 SolReasoning | OpenAI | 55 | 1M | 79/s | Paid |
| Claude Sonnet 5.5Reasoning | Anthropic | 54 | 1M | 138/s | Paid |
| GPT-6.1 TerraReasoning | OpenAI | 51 | 1M | 91/s | Paid |
| Claude Fable 5.1Reasoning | Anthropic | 50 | 1M | 64/s | Paid |
| Gemini 4 ArgonReasoning | 50 | 1M | 83/s | Paid | |
| Kimi K3.1 ProReasoning | Moonshot AI | 46 | 1M | 35/s | Paid |
| Muse WonderReasoning | Meta | 43 | 1.05M | 146/s | Paid |
| Claude Haiku 5.5Reasoning | Anthropic | 36 | 1M | 236/s | Paid |
| Grok 4.7Reasoning | xAI | 35 | 500K | 86/s | Paid |
| Qwen3.8 MaxReasoning | Alibaba | 34 | 984K | 40/s | Paid |
| DeepSeek V4.1 FlashReasoning, open | DeepSeek | 32 | 1M | 179/s | Paid |
| GLM-5.3Reasoning, open | Z.ai (Zhipu) | 32 | 1M | 72/s | Paid |
| Gemini 3.8 FlashReasoning | 31 | 1M | 129/s | Free | |
| MiniMax M3.1 FlashReasoning, open | MiniMax | 29 | 1M | 118/s | Paid |
| GLM-5.3 FlashReasoning, open | Z.ai (Zhipu) | 28 | 1M | 71/s | Paid |
| GPT-6 LunaChat | OpenAI | 28 | 1M | 118/s | Paid |
| Mistral Large 4Reasoning | Mistral | 26 | 1M | 161/s | Paid |
| Muse Spark OpenReasoning, open | Meta | 25 | 1.05M | 140/s | Paid |
| DeepSeek V4 ProReasoning, open | DeepSeek | 23 | 1M | 74/s | Paid |
| Qwen3.8 27BReasoning, open | Alibaba | 20 | 256K | 36/s | Free |
| MiniMax M3Reasoning, open | MiniMax | 19 | 1M | 103/s | Paid |
| Nemotron 3.5 Super VLReads images, open | NVIDIA | 14 | 1M | 444/s | Free |
| Gemini 3.5 Flash-LiteChat | 12 | 1M | 439/s | Free | |
| Nemotron 3 UltraReasoning, open | NVIDIA | 12 | 262K | 158/s | Free |
| Command A+Reads images | Cohere | 6 | 192K | 158/s | Free |
| Nemotron 3.5 LightningChat, open | NVIDIA | 6 | 1M | 293/s | Free |
| Mistral SmallChat | Mistral | 4 | 256K | 177/s | Free |
| CodestralCode | Mistral | Not scored | n/a | n/a | Free |
| Command AChat | Cohere | Not scored | 256K | 18/s | Free |
| Command A ReasoningReasoning | Cohere | Not scored | n/a | n/a | Free |
| Command A VisionReads images | Cohere | Not scored | n/a | n/a | Free |
| Command R7BChat | Cohere | Not scored | n/a | n/a | Free |
| DeepSeek R1Reasoning, open | DeepSeek | Not scored | n/a | n/a | Paid |
| DeepSeek V3.2Chat, open | DeepSeek | Not scored | n/a | n/a | Free |
| Gemini 3.1 Flash-LiteChat | Not scored | n/a | n/a | Free | |
| Gemini 3.1 ProReasoning | Not scored | 1M | 138/s | Paid | |
| Gemini 3.6 FlashReasoning | Not scored | n/a | n/a | Free | |
| Gemma 4 31BChat, open | Not scored | 256K | 36/s | Free | |
| GLM-4 9BChat, open | Z.ai (Zhipu) | Not scored | n/a | n/a | Free |
| GLM-4.5 FlashChat | Z.ai (Zhipu) | Not scored | n/a | n/a | Free |
| GLM-4.6V FlashReads images | Z.ai (Zhipu) | Not scored | n/a | n/a | Free |
| GLM-4.7 FlashChat | Z.ai (Zhipu) | Not scored | n/a | n/a | Free |
| GPT-6 SolReasoning | OpenAI | Not scored | n/a | n/a | Paid |
| GPT-OSS 120BReasoning, open | OpenAI | Not scored | 131K | 247/s | Free |
| Grok 4.20 Non-ReasoningChat | xAI | Not scored | n/a | n/a | Paid |
| Grok 4.20 ReasoningReasoning | xAI | Not scored | n/a | n/a | Paid |
| Grok 4.6Reasoning | xAI | Not scored | n/a | n/a | Paid |
| Kimi K2.6Chat, open | Moonshot AI | Not scored | n/a | n/a | Paid |
| Kimi K2.7 CodeCode | Moonshot AI | Not scored | 256K | 54/s | Paid |
| Kimi K3Reasoning, open | Moonshot AI | Not scored | n/a | n/a | Paid |
| Llama 3.3 70BChat, open | Meta | Not scored | 128K | 90/s | Free |
| Llama 4 MaverickChat, open | Meta | Not scored | 1M | 51/s | Paid |
| Llama Nemotron UltraReasoning, open | NVIDIA | Not scored | 128K | n/a | Free |
| MiniMax M2.7Reasoning, open | MiniMax | Not scored | n/a | n/a | Paid |
| Mistral LargeChat | Mistral | Not scored | n/a | n/a | Free |
| Mistral MediumChat | Mistral | Not scored | n/a | n/a | Free |
| Muse Spark 1.2Chat | Meta | Not scored | n/a | n/a | Paid |
| Muse Spark 1.3Chat | Meta | Not scored | n/a | n/a | Paid |
| Nemotron 3 SuperReasoning, open | NVIDIA | Not scored | n/a | n/a | Free |
| Phi-4Chat, open | Microsoft | Not scored | n/a | n/a | Paid |
| Qwen3 8BChat, open | Alibaba | Not scored | n/a | n/a | Free |
| Qwen3.7 PlusChat | Alibaba | Not scored | 1M | 57/s | Paid |
| Qwen3.8 FlashChat | Alibaba | Not scored | n/a | n/a | Paid |
| Qwen3.8 Omni FlashReads images | Alibaba | Not scored | n/a | n/a | Paid |
| SonarWeb search | Perplexity | Not scored | n/a | n/a | Paid |
| Sonar Deep ResearchWeb search | Perplexity | Not scored | n/a | n/a | Paid |
| Sonar ProWeb search | Perplexity | Not scored | n/a | n/a | Paid |
| Sonar Reasoning ProWeb search | Perplexity | Not scored | n/a | n/a | Paid |
- Gemini 4 Carbon59
Google, reasoning
Context n/aSpeed n/aPaid
- Claude Opus 5.558
Anthropic, reasoning
Context 1M97 tokens/sPaid
- GPT-6 Astra57
OpenAI, reasoning
Context 1M76 tokens/sPaid
- GPT-6.1 Sol55
OpenAI, reasoning
Context 1M79 tokens/sPaid
- Claude Sonnet 5.554
Anthropic, reasoning
Context 1M138 tokens/sPaid
- GPT-6.1 Terra51
OpenAI, reasoning
Context 1M91 tokens/sPaid
- Claude Fable 5.150
Anthropic, reasoning
Context 1M64 tokens/sPaid
- Gemini 4 Argon50
Google, reasoning
Context 1M83 tokens/sPaid
- Kimi K3.1 Pro46
Moonshot AI, reasoning
Context 1M35 tokens/sPaid
- Muse Wonder43
Meta, reasoning
Context 1.05M146 tokens/sPaid
- Claude Haiku 5.536
Anthropic, reasoning
Context 1M236 tokens/sPaid
- Grok 4.735
xAI, reasoning
Context 500K86 tokens/sPaid
- Qwen3.8 Max34
Alibaba, reasoning
Context 984K40 tokens/sPaid
- DeepSeek V4.1 Flash32
DeepSeek, reasoning, open
Context 1M179 tokens/sPaid
- GLM-5.332
Z.ai (Zhipu), reasoning, open
Context 1M72 tokens/sPaid
- Gemini 3.8 Flash31
Google, reasoning
Context 1M129 tokens/sFree API
- MiniMax M3.1 Flash29
MiniMax, reasoning, open
Context 1M118 tokens/sPaid
- GLM-5.3 Flash28
Z.ai (Zhipu), reasoning, open
Context 1M71 tokens/sPaid
- GPT-6 Luna28
OpenAI, chat
Context 1M118 tokens/sPaid
- Mistral Large 426
Mistral, reasoning
Context 1M161 tokens/sPaid
- Muse Spark Open25
Meta, reasoning, open
Context 1.05M140 tokens/sPaid
- DeepSeek V4 Pro23
DeepSeek, reasoning, open
Context 1M74 tokens/sPaid
- Qwen3.8 27B20
Alibaba, reasoning, open
Context 256K36 tokens/sFree API
- MiniMax M319
MiniMax, reasoning, open
Context 1M103 tokens/sPaid
- Nemotron 3.5 Super VL14
NVIDIA, reads images, open
Context 1M444 tokens/sFree API
- Gemini 3.5 Flash-Lite12
Google, chat
Context 1M439 tokens/sFree API
- Nemotron 3 Ultra12
NVIDIA, reasoning, open
Context 262K158 tokens/sFree API
- Command A+6
Cohere, reads images
Context 192K158 tokens/sFree API
- Nemotron 3.5 Lightning6
NVIDIA, chat, open
Context 1M293 tokens/sFree API
- Mistral Small4
Mistral, chat
Context 256K177 tokens/sFree API
- CodestralNot scored
Mistral, code
Context n/aSpeed n/aFree API
- Command ANot scored
Cohere, chat
Context 256K18 tokens/sFree API
- Command A ReasoningNot scored
Cohere, reasoning
Context n/aSpeed n/aFree API
- Command A VisionNot scored
Cohere, reads images
Context n/aSpeed n/aFree API
- Command R7BNot scored
Cohere, chat
Context n/aSpeed n/aFree API
- DeepSeek R1Not scored
DeepSeek, reasoning, open
Context n/aSpeed n/aPaid
- DeepSeek V3.2Not scored
DeepSeek, chat, open
Context n/aSpeed n/aFree API
- Gemini 3.1 Flash-LiteNot scored
Google, chat
Context n/aSpeed n/aFree API
- Gemini 3.1 ProNot scored
Google, reasoning
Context 1M138 tokens/sPaid
- Gemini 3.6 FlashNot scored
Google, reasoning
Context n/aSpeed n/aFree API
- Gemma 4 31BNot scored
Google, chat, open
Context 256K36 tokens/sFree API
- GLM-4 9BNot scored
Z.ai (Zhipu), chat, open
Context n/aSpeed n/aFree API
- GLM-4.5 FlashNot scored
Z.ai (Zhipu), chat
Context n/aSpeed n/aFree API
- GLM-4.6V FlashNot scored
Z.ai (Zhipu), reads images
Context n/aSpeed n/aFree API
- GLM-4.7 FlashNot scored
Z.ai (Zhipu), chat
Context n/aSpeed n/aFree API
- GPT-6 SolNot scored
OpenAI, reasoning
Context n/aSpeed n/aPaid
- GPT-OSS 120BNot scored
OpenAI, reasoning, open
Context 131K247 tokens/sFree API
- Grok 4.20 Non-ReasoningNot scored
xAI, chat
Context n/aSpeed n/aPaid
- Grok 4.20 ReasoningNot scored
xAI, reasoning
Context n/aSpeed n/aPaid
- Grok 4.6Not scored
xAI, reasoning
Context n/aSpeed n/aPaid
- Kimi K2.6Not scored
Moonshot AI, chat, open
Context n/aSpeed n/aPaid
- Kimi K2.7 CodeNot scored
Moonshot AI, code
Context 256K54 tokens/sPaid
- Kimi K3Not scored
Moonshot AI, reasoning, open
Context n/aSpeed n/aPaid
- Llama 3.3 70BNot scored
Meta, chat, open
Context 128K90 tokens/sFree API
- Llama 4 MaverickNot scored
Meta, chat, open
Context 1M51 tokens/sPaid
- Llama Nemotron UltraNot scored
NVIDIA, reasoning, open
Context 128KSpeed n/aFree API
- MiniMax M2.7Not scored
MiniMax, reasoning, open
Context n/aSpeed n/aPaid
- Mistral LargeNot scored
Mistral, chat
Context n/aSpeed n/aFree API
- Mistral MediumNot scored
Mistral, chat
Context n/aSpeed n/aFree API
- Muse Spark 1.2Not scored
Meta, chat
Context n/aSpeed n/aPaid
- Muse Spark 1.3Not scored
Meta, chat
Context n/aSpeed n/aPaid
- Nemotron 3 SuperNot scored
NVIDIA, reasoning, open
Context n/aSpeed n/aFree API
- Phi-4Not scored
Microsoft, chat, open
Context n/aSpeed n/aPaid
- Qwen3 8BNot scored
Alibaba, chat, open
Context n/aSpeed n/aFree API
- Qwen3.7 PlusNot scored
Alibaba, chat
Context 1M57 tokens/sPaid
- Qwen3.8 FlashNot scored
Alibaba, chat
Context n/aSpeed n/aPaid
- Qwen3.8 Omni FlashNot scored
Alibaba, reads images
Context n/aSpeed n/aPaid
- SonarNot scored
Perplexity, web search
Context n/aSpeed n/aPaid
- Sonar Deep ResearchNot scored
Perplexity, web search
Context n/aSpeed n/aPaid
- Sonar ProNot scored
Perplexity, web search
Context n/aSpeed n/aPaid
- Sonar Reasoning ProNot scored
Perplexity, web search
Context n/aSpeed n/aPaid
Speed is output tokens a second. Context is the most text a model reads at once. Free means at least one host gives a free key that reaches it.
Video models
Describe a scene and these make a clip; it plays in its panel with a Save button. Write Part 1, Part 2 and more, and the models marked below carry each part on from the last frame of the one before, so the parts play as one video.
| Model | Clip length | Resolution | Sound | Parts carry on | Where |
|---|---|---|---|---|---|
| Gemini Omni FlashGoogle | Not listed | Not listed | Not listed | No | |
| Grok Imagine Video 1.5xAI | Up to 15 s | Not listed | Not listed | No | xAI |
| Grok Imagine Video 1.5 LitexAI | Up to 15 s | Not listed | Not listed | No | xAI |
| Seedance 1.5 proByteDance | 4 to 12 s | Up to 1080p | Yes | Yes | BytePlus ModelArk |
| Seedance 2.0ByteDance | 4 to 15 s | Up to 4K | Yes | Yes | BytePlus ModelArk |
| Seedance 2.0 FastByteDance | 4 to 15 s | Up to 720p | Yes | Yes | BytePlus ModelArk |
| Seedance 2.5ByteDance | 4 to 30 s | Up to 1080p | Yes | Yes | BytePlus ModelArk |
From each maker's API docs, 30 October 2026. Not listed means the docs did not say.
Why pick one? Use them all at once
A benchmark tells you how a model does on average. MultiScreenAI shows how they do on your own question: one message goes to every model on one screen, and the blind test mode lets you judge the answers without knowing who wrote them.