AInnect

AI model leaderboards

Frontier and open-weight model rankings, what developers actually run, and the most-downloaded open video models. Each board links to its source.

Frontier models

Blind head-to-head votes from real users, aggregated into an Elo rating. Higher means the model won more pairwise comparisons.

#ModelOrgElo
1Claude Fable 5Anthropic1507
2Claude Opus 4.6-highAnthropic1505
3Muse Spark 1.2Meta1498
4Gemini 3.1 ProGoogle1487
5GPT-5.6 SolOpenAI1481
6Qwen3.8-MaxAlibaba1479
7Grok 4.6xAI1461

Elo is relative — a 20-point gap is small; only large gaps are meaningful.

Source: LMArena text arena ↗

Open-weight models

Open-weight models you can download and self-host, scored on a 0–100 composite index.

#ModelOrgindex
1Qwen3.8-MaxAlibaba79
2Kimi K3Moonshot60
3GLM-5.3Z.ai58
4DeepSeek V4DeepSeek58
5Llama 4Meta52

Scored 0–100. Not comparable with the Elo ratings on the frontier board.

Source: Open-weights rankings ↗

Most used on OpenRouter

What developers actually run: models ranked by tokens processed through OpenRouter over the last day.

#ModelOrgT tokens
1deepseek-v4-flashdeepseek12.19
2ox-alphastealth9.75
3gpt-5.6-lunaopenai8.48
4glm-5.3-flashz-ai8.14
5mimo-v2.5xiaomi8.08
6hy3tencent6.41
7nemotron-3-ultra-550b-a55bnvidia4.93
8hy4-previewtencent4.01
9gemini-3.7-flashgoogle3.85
10minimax-m3minimax3.02

Trillions of tokens processed. Usage, not quality — cheap and free models rank high here.

Source: OpenRouter ↗

Open-weight video models

Open-weight video generators you can download and run, ranked by downloads across every variant of each model.

#ModelOrgM downloads
1Wan2.2Wan-AI6.48
2MiniMax-H3larryvrh4.02
3LTX-2.3Lightricks2.57
4Wan2.1Wan-AI2.07
5LTX-2.5Lightricks1.49
6LTX-VideoLightricks0.5
7LTX-2Lightricks0.38
8Sulphur-2SulphurAI0.35
9stable-video-diffusion-img2vidstabilityai0.25
10Cosmos-Predict2nvidia0.18

Millions of downloads, summing quantised and repackaged variants into their base model.

Source: Hugging Face ↗