Live model intelligence map

Explore LLMs by the tradeoffs that decide production choices

Compare frontier and open-weight models by intelligence, agentic capability, cost per task, token price, latency, throughput, context, and provider availability.

Live350 models87 providersArtificial Analysis Intelligence Index v4.1Updated June 2026
Decision cockpit
206 model groups654 endpointsAA v4.1

Choose by task economics, not leaderboard position alone.

Explore now brings Artificial Analysis v4.1 task metrics into the main model map: agentic strength, per-task cost, response time, context, provider coverage, and token price in one view.

Model groups

Showing 206 rows after filters.

Action
GLM-5.2Open

Z AI

51.143.1$0.376$1.34/1M6.0s1.0M16Compare
Muse Spark 1.1 (xhigh)Open

Meta

50.637.5$0.261$2.00/1M22s1.0M1Compare
Qwen3.7 MaxOpen

Alibaba

4630.6$0.985$1.88/1M17s1.0M2Compare
MiniMax-M3Open

MiniMax

44.435.4$0.125$0.525/1M24s1.0M5Compare
DeepSeek V4 ProOpen

DeepSeek

44.336.4$0.045$0.544/1M9.2s1.0M10Compare
Kimi K2.6Open

Kimi

44.230.3N/A$1.40/1M13s262K13Compare
MiMo-V2.5-ProOpen

Xiaomi

42.229.1$0.031$0.544/1M9.8s1.0M3Compare
Kimi K2.7 CodeOpen

Kimi

41.929.6N/A$1.43/1M57s262K11Compare
Nex-N2-ProOpen

Nex AGI

4131N/A$1.00/1M22s262K1Compare
DeepSeek V4 FlashOpen

DeepSeek

40.331.1$0.022$0.113/1M6.0s1.0M7Compare
GLM-5.1 (Reasoning)Open

Z AI

40.229.9$0.247$1.55/1M54s205K9Compare
Qwen3.6 Max PreviewOpen

Alibaba

40N/AN/A$2.92/1M50s256K1Compare
Qwen3.6 PlusOpen

Alibaba

39.627.6$0.314$1.13/1M2.0 min1.0M1Compare
GLM-5 (Reasoning)Open

Z AI

39.5N/AN/A$0.970/1M71s203K4Compare
Qwen3.7 PlusOpen

Alibaba

3920.8$0.195$0.700/1M52s1.0M1Compare
MiniMax-M2.7Open

MiniMax

38.125.6N/A$0.525/1M54s205K5Compare
Kimi K2.5 (Reasoning)Open

Kimi

38.1N/AN/A$0.900/1M64s262K7Compare
Nemotron 3 Ultra 550B A55B (Reasoning)Open

NVIDIA

37.827.4$0.234$0.547/1M15s512K6Compare
MiMo-V2.5Open

Xiaomi

37.223.7$0.0098$0.175/1M31s1.0M4Compare
Qwen3.6 27B (Reasoning)Open

Alibaba

37.127$0.265$1.02/1M1.8 min262K6Compare
GLM-5.1Open

Z AI

35.4N/AN/A$1.55/1M9.2s205K7Compare
Qwen3.5 27B (Reasoning)Open

Alibaba

33.8N/AN/A$0.688/1M37s262K5Compare
Qwen3.5 397B A17B (Reasoning)Open

Alibaba

33.719.8$0.333$0.877/1M61s262K8Compare
MiniMax-M2.5Open

MiniMax

33.7N/AN/A$0.394/1M32s205K10Compare
GLM-4.7 (Reasoning)Open

Z AI

33.725.4$0.297$0.738/1M19s205K6Compare
Hy3Open

Tencent

33.6N/AN/A$0.0000/1M18s262K1Compare
Hy3-preview (Reasoning)Open

Tencent

33.6N/AN/A$0.115/1M18s256K1Compare
Kimi K2 ThinkingOpen

Kimi

32.7N/AN/A$1.07/1M16s262K3Compare
GLM-5Open

Z AI

32.4N/AN/A$0.970/1M8.6s203K3Compare
Qwen3.5 122B A10B (Reasoning)Open

Alibaba

32.320.7$0.241$0.715/1M20s262K4Compare
Qwen3.5 397B A17BOpen

Alibaba

32N/AN/A$0.901/1M11s262K6Compare
DeepSeek V3.2 (Reasoning)Open

DeepSeek

3218.3N/A$0.302/1MN/A164K4Compare
Qwen3.6 35B A3B (Reasoning)Open

Alibaba

31.621.4$0.179$0.350/1M36s1.0M8Compare
MiniMax-M2.1Open

MiniMax

31.4N/AN/A$0.525/1M30s205K2Compare
Qwen3.5 Omni PlusOpen

Alibaba

30.6N/AN/A$1.50/1M12s256K1Compare
Ring-2.6-1TOpen

InclusionAI

30.618.9$0.345$0.850/1M23s262K1Compare
Qwen3.6 27BOpen

Alibaba

30.523.3$0.359$1.02/1M13s262K5Compare
DeepSeek V3.1 Terminus (Reasoning)Open

DeepSeek

30.418.1N/A$0.453/1MN/A131K1Compare
Step 3.7 FlashOpen

StepFun

30.321.5N/A$0.438/1M7.3s256K2Compare
Mistral Medium 3.5Open

Mistral

29.919$1.20$3.00/1M19s262K1Compare
Gemma 4 31B (Reasoning)Open

Google

29.414.4N/A$0.0000/1M65s262K12Compare
Kimi K2.5Open

Kimi

29.4N/AN/A$1.20/1M12s262K4Compare
Qwen3.5 35B A3B (Reasoning)Open

Alibaba

29.3N/AN/A$0.355/1M17s262K4Compare
Qwen3.5 27BOpen

Alibaba

29.3N/AN/A$0.825/1M11s262K3Compare
GLM-4.6 (Reasoning)Open

Z AI

28.717.7$0.283$0.758/1M53s205K2Compare
MiniMax-M2Open

MiniMax

28.3N/AN/A$0.525/1M30s205K4Compare
Qwen3.5 122B A10BOpen

Alibaba

27.615.8$0.177$0.817/1M5.6s262K2Compare
GLM-4.7Open

Z AI

26.6N/AN/A$0.738/1M4.4s205K6Compare
Hy3-previewOpen

Tencent

26.1N/AN/A$0.115/1M6.1s328K1Compare
Gemma 4 26B A4B (Reasoning)Open

Google

25.711$0.029$0.0000/1MN/A1.0M6Compare
Step 3.5 FlashOpen

StepFun

25.5N/AN/A$0.150/1M13s256K1Compare
NVIDIA Nemotron 3 Super 120B A12B (Reasoning)Open

NVIDIA

25.48.7$0.205$0.350/1M12s262K2Compare
DeepSeek V3.2 Exp (Reasoning)Open

DeepSeek

25.4N/AN/A$0.305/1MN/A164K1Compare
Qwen3 Max ThinkingOpen

Alibaba

25N/AN/A$2.40/1M45s262K1Compare
DeepSeek V3.2Open

DeepSeek

24.7N/AN/A$0.290/1MN/A164K9Compare
Trinity Large ThinkingOpen

Arcee AI

24.5N/AN/A$0.378/1M13s262K2Compare
Qwen3.6 35B A3BOpen

Alibaba

24.2N/AN/A$0.350/1M5.0s1.0M7Compare
Qwen3 MaxOpen

Alibaba

24N/AN/A$2.40/1M12s262K2Compare
gpt-oss-120b (high)Open

OpenAI

23.813.2$0.061$0.065/1M8.6s131K18Compare
Kimi K2 0905Open

Kimi

23.5N/AN/A$1.07/1M16s262K1Compare
Qwen3.5 35B A3BOpen

Alibaba

23.4N/AN/A$0.355/1M4.9s262K2Compare
GLM-4.6Open

Z AI

23N/AN/A$0.963/1M15s205K1Compare
GLM-4.7-Flash (Reasoning)Open

Z AI

22.9N/AN/A$0.145/1M26s203K3Compare
Command A+Open

Cohere

22.59.2$0.0000$0.0000/1M14s200K1Compare
Gemma 4 12B (Reasoning)Open

Google

22N/AN/A$0.150/1M26s262K1Compare
Gemma 4 31BOpen

Google

21.811$0.034$0.193/1M12s262K8Compare
Qwen3.5 9B (Reasoning)Open

Alibaba

21.47.4$0.164$0.113/1M48s262K2Compare
DeepSeek V3.1 TerminusOpen

DeepSeek

21.4N/AN/A$0.440/1MN/A164K2Compare
DeepSeek V3.2 ExpOpen

DeepSeek

21.3N/AN/A$0.305/1MN/A164K1Compare
Qwen3 Coder NextOpen

Alibaba

21.18.8$0.329$0.313/1M6.6s262K3Compare
DeepSeek V3.1Open

DeepSeek

21N/AN/A$0.355/1MN/A164K7Compare
DeepSeek V3.1 (Reasoning)Open

DeepSeek

20.7N/AN/A$0.453/1MN/A164K3Compare
North Mini CodeOpen

Cohere

20.6N/AN/A$0.0000/1M23s256K1Compare
Qwen3 VL 235B A22B (Reasoning)Open

Alibaba

20.6N/AN/A$1.72/1M48s131K2Compare
Gemma 4 26B A4BOpen

Google

20.1N/AN/A$0.138/1M6.8s1.0M6Compare
Qwen3.5 4B (Reasoning)Open

Alibaba

20.1N/AN/A$0.060/1M1.5 min262K1Compare
DeepSeek R1 0528Open

DeepSeek

20.1N/AN/A$0.912/1MN/A164K4Compare
Mistral Small 4 (Reasoning)Open

Mistral

19.64.7N/A$0.262/1M16s256K1Compare
Qwen3 235B A22B 2507 (Reasoning)Open

Alibaba

19.63.8N/A$0.100/1M47s262K4Compare
Kimi K2Open

Kimi

19.4N/AN/A$1.00/1M16s131K1Compare
Devstral 2Open

Mistral

19.210.6$0.0000$0.0000/1M8.6s262K1Compare
Qwen3.5 Omni FlashOpen

Alibaba

19N/AN/A$0.275/1M4.1s256K1Compare
DeepSeek R1Open

DeepSeek

18.53.1$0.247$1.15/1MN/A128K3Compare
Seed-OSS-36B-InstructOpen

ByteDance Seed

18.3N/AN/A$0.300/1M73s262K1Compare
Qwen3 235B A22B 2507 InstructOpen

Alibaba

18.2N/AN/A$0.092/1M11s262K10Compare
Qwen3 Coder 480B A35B InstructOpen

Alibaba

18N/AN/A$0.475/1M11s262K6Compare
Magistral Medium 1.2Open

Mistral

17.91.9$0.747$2.75/1M64s131K1Compare
Qwen3 VL 32B (Reasoning)Open

Alibaba

17.9N/AN/A$2.63/1M31s131K1Compare
HyperNova 60B 2605Open

Multiverse Computing

17.86.7$0.018$0.065/1M8.0s131K1Compare
gpt-oss-120b (low)Open

OpenAI

17.7N/AN/A$0.065/1M8.4s131K16Compare
MiniMax M1 80kOpen

MiniMax

17.7N/AN/A$0.963/1MN/A1.0M1Compare
Devstral Small 2Open

Mistral

17.410.5$0.0000$0.0000/1M8.0s256K1Compare
GLM-4.6V (Reasoning)Open

Z AI

16.8N/AN/A$0.450/1M49s131K1Compare
Qwen3 Next 80B A3B (Reasoning)Open

Alibaba

16.72.1$0.169$0.412/1M16s262K3Compare
GLM-4.5-AirOpen

Z AI

16.5N/AN/A$0.320/1M38s98K1Compare
Qwen3.5 4BOpen

Alibaba

16N/AN/A$0.060/1M15s262K1Compare
Mistral Large 3Open

Mistral

15.95.5$0.060$0.750/1M11s262K3Compare
GLM-4.7-FlashOpen

Z AI

15.5N/AN/A$0.153/1M4.2s200K2Compare
DeepSeek V3 0324Open

DeepSeek

15.41.5$0.190$0.405/1MN/A164K3Compare
gpt-oss-20b (high)Open

OpenAI

14.93.1$0.018$0.055/1M11s131K10Compare
Nemotron 3 Nano Omni 30B A3B ReasoningOpen

NVIDIA

14.9N/AN/A$0.105/1M8.8s66K1Compare
Mistral Small 3.1Open

Mistral

14.75.2$0.041$0.063/1M4.0s131K4Compare
Mistral Medium 3.1Open

Mistral

14.76.2$0.145$0.800/1M7.7s131K1Compare
Qwen3 30B A3B 2507 (Reasoning)Open

Alibaba

14.41.8$0.072$0.750/1M24s262K1Compare
gpt-oss-20b (low)Open

OpenAI

14.3N/AN/A$0.055/1M10s131K9Compare
Llama 4 MaverickOpen

Meta

14.31.3$0.032$0.262/1M5.2s1.0M7Compare
Qwen3 VL 235B A22B InstructOpen

Alibaba

14.3N/AN/A$0.600/1M13s131K3Compare
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)Open

NVIDIA

14.22$0.018$0.088/1M27s262K3Compare
DeepSeek V3Open

DeepSeek

14.21.6$0.025$0.463/1MN/A164K2Compare
Ling 2.6 FlashOpen

InclusionAI

14.12.3N/A$0.150/1M4.3s262K1Compare
Qwen3 Next 80B A3B InstructOpen

Alibaba

13.7N/AN/A$0.343/1M5.0s262K6Compare
Qwen3 Coder 30B A3B InstructOpen

Alibaba

13.6N/AN/A$0.262/1M7.8s262K3Compare
Qwen3 235B A22B (Reasoning)Open

Alibaba

13.4N/AN/A$2.63/1M43s131K1Compare
QwQ 32BOpen

Alibaba

13.4N/AN/A$0.745/1M1.6 min24K1Compare
Qwen3 VL 30B A3B (Reasoning)Open

Alibaba

13.3N/AN/A$0.750/1M25s131K1Compare
Gemma 4 12BOpen

Google

13.2N/AN/A$0.150/1M7.3s262K1Compare
Mistral Medium 3Open

Mistral

12.5N/AN/A$0.800/1M13s131K2Compare
Mistral Small 4Open

Mistral

12.4N/AN/A$0.262/1M4.1s262K1Compare
Llama Nemotron Super 49B v1.5 (Reasoning)Open

NVIDIA

12.4N/AN/A$0.400/1M52s131K1Compare
Sarvam 105B (high)Open

Sarvam

11.9N/AN/A$0.074/1M24s128K1Compare
Showing 120 of 206 model groups.

How to use this LLM ranking page

Start with capability, then force the economics

Frontier models often cluster tightly on raw intelligence. Cost per task, response time, provider coverage, and context length usually create the real shortlist.

Move from discovery to a defensible final pick

Use Explore to find the shape of the market, then move into Compare or use the Agentic Fit Finder when task-level reliability and cost matter more than chat quality alone.

Frequently Asked Questions

What is the best LLM in 2026?

Claude Fable 5 leads the overall quality ranking right now. The best model for you depends on your use case — coding, cost, speed, and context length all shift the answer.

How do I compare LLMs?

Sort by Quality Index for overall strength, then filter by price, speed, or context window to match your constraints. Move to the Compare page to put 2–4 finalists head to head.

Which AI model has the best quality-to-price ratio?

Open-weight models like DeepSeek, Qwen, and Llama often lead on quality-to-price. For agentic workflows, cost per task is usually a better filter than token price alone.

How often is this leaderboard updated?

Data is pulled from Artificial Analysis and refreshed automatically. New models appear as soon as they have benchmark scores and provider endpoints.