Model rankings
1319 models · 5 boards · 2 sources
Smartest
Best open model
Best value
Best from China
Top three by Intelligence
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic55.7
GPT-5.5 (xhigh)OpenAI54.8
Top three open models
GLM-5.2 (max)Z AI · #7 overall51.1
MiniMax-M3MiniMax · #12 overall44.4
DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek · #13 overall44.3
This week's changes
Reve 2.0Image · #2New
Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Intelligence · #5New
claude-sonnet-5-thinkingCoding · #5New
Fish Audio S2.1 ProSpeech · #15New
OpenAudio S1 MiniSpeech · #50New
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | — | Nemotron 3 Ultra 550B A55B (Reasoning)OpenNVIDIA | 37.8 | $1.2 | 132 | 1.2 s |
| 2 | — | Gemma 4 31B (Reasoning)OpenGoogle · 1 more variant | 29.4 | $0.000 | 35 | 1.1 s |
| 3 | — | Gemma 4 26B A4B (Reasoning)OpenGoogle · 1 more variant | 25.7 | $0.20 | — | — |
| 4 | — | NVIDIA Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA | 25.4 | $0.41 | 223 | 1.4 s |
| 5 | — | gpt-oss-120b (high)OpenOpenAI · 1 more variant | 23.8 | $0.26 | 261 | 1.0 s |
| 6 | — | Gemma 4 12B (Reasoning)OpenGoogle · 1 more variant | 22.0 | $0.15 | 120 | 2.4 s |
| 7 | — | Nemotron Cascade 2 30B A3BOpenNVIDIA | 21.3 | — | — | — |
| 8 | New | Apriel-v1.5-15B-ThinkerOpenServiceNow | 21.2 | $0.000 | — | — |
| 9 | New | Apriel-v1.6-15B-ThinkerOpenServiceNow | 20.5 | $0.000 | — | — |
| 10 | ▼ 2 | Nemotron 3 Nano Omni 30B A3B ReasoningOpenNVIDIA | 14.9 | $0.13 | 292 | 1.0 s |
| 11 | ▼ 2 | gpt-oss-20b (high)OpenOpenAI · 1 more variant | 14.9 | $0.088 | 234 | 0.8 s |
| 12 | ▼ 2 | Llama 4 MaverickOpenMeta | 14.3 | $0.48 | 107 | 1.0 s |
| 13 | ▼ 2 | NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)OpenNVIDIA · 1 more variant | 14.2 | $0.096 | 96 | 2.6 s |
| 14 | New | DiffusionGemma 26B A4BOpenGoogle | 13.5 | — | — | — |
| 15 | ▼ 3 | Gemma 4 E4B (Reasoning)OpenGoogle · 1 more variant | 12.5 | — | — | — |
| 16 | ▼ 3 | Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant | 12.4 | $0.18 | 50 | 1.3 s |
| 17 | ▼ 3 | Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant | 12.2 | — | — | — |
| 18 | ▼ 3 | Llama 4 ScoutOpenMeta | 10.0 | $0.29 | 105 | 0.9 s |
| 19 | ▼ 3 | Gemma 4 E2B (Reasoning)OpenGoogle · 1 more variant | 9.3 | — | — | — |
| 20 | ▼ 3 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA | 9.1 | $0.90 | 52 | 2.4 s |
| 21 | ▼ 3 | NVIDIA Nemotron Nano 12B v2 VL (Reasoning)OpenNVIDIA · 1 more variant | 9.0 | $0.30 | 290 | 0.4 s |
| 22 | ▼ 3 | Granite 4.1 30BOpenIBM | 8.9 | — | — | — |
| 23 | ▼ 3 | NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA · 1 more variant | 8.8 | $0.070 | 84 | 8.6 s |
| 24 | ▼ 3 | NVIDIA Nemotron 3 Nano 4BOpenNVIDIA | 8.8 | — | — | — |
| 25 | ▼ 3 | Llama 3.3 Instruct 70BOpenMeta | 8.6 | $0.61 | 91 | 1.6 s |
| 26 | ▼ 3 | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA | 8.5 | — | — | — |
| 27 | ▼ 3 | Llama 3.1 Instruct 405BOpenMeta | 8.5 | $3.7 | 75 | 2.4 s |
| 28 | ▼ 3 | LFM2.5-8B-A1BOpenLiquid AI | 8.3 | $0.000 | 340 | 2.6 s |
| 29 | ▼ 3 | Grok 2 (Dec '24)OpenxAI | 8.0 | — | — | — |
| 30 | ▼ 3 | Llama 3.1 Nemotron Instruct 70BOpenNVIDIA | 7.6 | $1.2 | 301 | 4.9 s |
| 31 | ▲ 1 | Llama 3.1 Instruct 8BOpenMeta | 7.6 | $0.080 | 141 | 0.9 s |
| 32 | ▲ 5 | Gemma 3 27B InstructOpenGoogle | 7.4 | $0.000 | — | — |
| 33 | ▼ 5 | Llama 3.1 Instruct 70BOpenMeta | 6.8 | $0.56 | 32 | 2.0 s |
| 34 | ▼ 5 | Granite 4.1 8BOpenIBM | 6.7 | $0.063 | 114 | 0.8 s |
| 35 | ▼ 5 | R1 1776OpenPerplexity | 6.3 | — | — | — |
| 36 | ▼ 5 | Llama 3.2 Instruct 90B (Vision)OpenMeta | 6.2 | $1.4 | 58 | 1.2 s |
| 37 | ▼ 4 | Grok-1OpenxAI | 6.0 | — | — | — |
| 38 | ▼ 4 | Granite 4.0 H SmallOpenIBM | 5.2 | $0.11 | 413 | 10 s |
| 39 | ▼ 4 | LFM2 24B A2BOpenLiquid AI | 4.9 | $0.053 | 120 | 0.6 s |
| 40 | ▼ 4 | Phi-4OpenMicrosoft | 4.9 | $0.22 | 36 | 2.2 s |
| 41 | ▼ 3 | Phi-3 Mini Instruct 3.8BOpenMicrosoft | 4.6 | — | — | — |
| 42 | ▼ 3 | Gemma 3n E4B Instruct Preview (May '25)OpenGoogle | 4.6 | — | — | — |
| 43 | ▼ 3 | Phi-4 Multimodal InstructOpenMicrosoft | 4.5 | $0.000 | 17 | 1.3 s |
| 44 | ▼ 3 | Llama 2 Chat 7BOpenMeta | 4.3 | $0.10 | — | — |
| 45 | ▼ 3 | Llama 3.2 Instruct 3BOpenMeta | 4.2 | $0.15 | 52 | 1.1 s |
| 46 | ▼ 3 | Llama 3 Instruct 70BOpenMeta | 3.5 | $1.2 | — | — |
| 47 | ▼ 3 | Arctic InstructOpenSnowflake | 3.4 | — | — | — |
| 48 | ▼ 3 | Gemma 3 12B InstructOpenGoogle | 3.4 | $0.000 | — | — |
| 49 | ▼ 3 | Llama 3.2 Instruct 11B (Vision)OpenMeta | 3.3 | $0.25 | 49 | 0.8 s |
| 50 | ▼ 3 | Granite 4.1 3BOpenIBM | 3.2 | — | — | — |
Intelligence over time
Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi