Tipping Point
Track AI capability domains approaching singularity thresholds — benchmark scores, leaderboard rankings, and projected crossing dates.
Last refreshedJul 17, 2026, 07:41 AMvia Live web search (0 source(s))
Capability Domains
Capability Growth
Latest News
Influencer claims Qwen 3.7 Max ends the model race
AI Influencer BlogJun 7
Viral post: Llama 5 makes paid APIs obsolete
X / TwitterMay 18
Experts refute viral AGI demo methodology
MIT Technology ReviewApr 28
DeepSeek V4 Pro reportedly aces the math olympiad
AI News DailyApr 28
Hype trend: boiling claims rising in 2026
MIT Technology ReviewApr 28
Singularity Timeline Projection
AGI Leaderboard
| Model | Vendor | Knowledge | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| #39 | Llama 5 | Meta | 91.0 | 0.9 | 0.8 | — | 0.9 | 1M | 140 t/s | 90.9 pts/$ |
| #38 | GLM-5.2 | Zhipu AI | 92.0 | 0.9 | 0.8 | — | 0.9 | 1M | 150 t/s | 60.0 pts/$ |
| #36 | Qwen 3.7 Max | Alibaba | 91.0 | 0.9 | 0.8 | — | 0.9 | 1M | 155 t/s | 65.0 pts/$ |
| #35 | Claude Sonnet 5 | Anthropic | 90.0 | 0.9 | 0.8 | — | 0.9 | 1M | 180 t/s | 45.2 pts/$ |
| #34 | Grok 4.5 | xAI | 93.0 | 0.9 | 0.9 | — | 0.9 | 1M | 175 t/s | 37.1 pts/$ |
| #33 | Gemini 3.1 Pro | Google DeepMind | 92.0 | 0.9 | 0.9 | — | — | 1M | 190 t/s | 30.7 pts/$ |
| #31 | GPT-5.5 | OpenAI | 93.0 | 0.9 | 0.9 | — | 0.9 | 1M | 160 t/s | 23.3 pts/$ |
| #29 | DeepSeek V4 Pro | DeepSeek | 93.0 | 0.9 | 0.9 | — | — | 1M | 130 t/s | 93.4 pts/$ |
| #28 | Claude Fable 5 | Anthropic | 95.0 | 1.0 | 0.9 | — | 0.9 | 1M | 120 t/s | 15.9 pts/$ |
| #25 | Gemini 3.2 Pro | Google DeepMind | 96.0 | 1.0 | 0.9 | — | 0.9 | 2M | 200 t/s | 19.2 pts/$ |
| #24 | Claude Opus 4.8 | Anthropic | 95.0 | 1.0 | 0.9 | 0.8 | — | 500K | 110 t/s | 18.9 pts/$ |
| #20 | Llama 5 | Meta | 91.0 | 0.9 | 0.8 | — | 0.9 | 1M | 140 t/s | 90.9 pts/$ |
| #20 | Llama 5 | Meta | 91.0 | 0.9 | 0.8 | — | 0.9 | 1M | 140 t/s | 90.9 pts/$ |
| #19 | Claude Sonnet 5 | Anthropic | 90.0 | 0.9 | 0.8 | — | 0.9 | 1M | 180 t/s | 45.2 pts/$ |
| #19 | Claude Sonnet 5 | Anthropic | 90.0 | 0.9 | 0.8 | — | 0.9 | 1M | 180 t/s | 45.2 pts/$ |
| #17 | Qwen 3.7 Max | Alibaba | 91.0 | 0.9 | 0.8 | — | 0.9 | 1M | 155 t/s | 65.0 pts/$ |
| #16 | GLM-5.2 | Zhipu AI | 92.0 | 0.9 | 0.8 | — | 0.9 | 1M | 150 t/s | 60.0 pts/$ |
| #16 | GLM-5.2 | Zhipu AI | 92.0 | 0.9 | 0.8 | — | 0.9 | 1M | 150 t/s | 60.0 pts/$ |
| #14 | DeepSeek V4 Pro | DeepSeek | 93.0 | 0.9 | 0.9 | — | — | 1M | 130 t/s | 93.4 pts/$ |
| #13 | Gemini 3.1 Pro | Google DeepMind | 92.0 | 0.9 | 0.9 | — | — | 1M | 190 t/s | 30.7 pts/$ |
| #11 | GPT-5.6 | OpenAI | 96.0 | 1.0 | 0.9 | 0.9 | 0.9 | 2M | 145 t/s | 12.8 pts/$ |
| #11 | GPT-5.6 | OpenAI | 96.0 | 1.0 | 0.9 | 0.9 | 0.9 | 2M | 145 t/s | 12.8 pts/$ |
| #10 | Qwen 3.7 Max | Alibaba | 91.0 | 0.9 | 0.8 | — | 0.9 | 1M | 155 t/s | 65.0 pts/$ |
| #9 | Grok 4.5 | xAI | 93.0 | 0.9 | 0.9 | — | 0.9 | 1M | 175 t/s | 37.1 pts/$ |
| #9 | Grok 4.5 | xAI | 93.0 | 0.9 | 0.9 | — | 0.9 | 1M | 175 t/s | 37.1 pts/$ |
| #8 | GPT-5.5 | OpenAI | 93.0 | 0.9 | 0.9 | — | 0.9 | 1M | 160 t/s | 23.3 pts/$ |
| #8 | GPT-5.5 | OpenAI | 93.0 | 0.9 | 0.9 | — | 0.9 | 1M | 160 t/s | 23.3 pts/$ |
| #7 | Claude Opus 4.8 | Anthropic | 95.0 | 1.0 | 0.9 | 0.8 | — | 500K | 110 t/s | 18.9 pts/$ |
| #7 | Claude Opus 4.8 | Anthropic | 95.0 | 1.0 | 0.9 | 0.8 | — | 500K | 110 t/s | 18.9 pts/$ |
| #6 | DeepSeek V4 Pro | DeepSeek | 93.0 | 0.9 | 0.9 | — | — | 1M | 130 t/s | 93.4 pts/$ |
| #5 | Claude Fable 5 | Anthropic | 95.0 | 1.0 | 0.9 | — | 0.9 | 1M | 120 t/s | 15.9 pts/$ |
| #5 | Claude Fable 5 | Anthropic | 95.0 | 1.0 | 0.9 | — | 0.9 | 1M | 120 t/s | 15.9 pts/$ |
| #4 | Claude Fable 5 | Anthropic | 95.0 | 1.0 | 0.9 | — | 0.9 | 1M | 120 t/s | 15.9 pts/$ |
| #3 | Gemini 3.2 Pro | Google DeepMind | 96.0 | 1.0 | 0.9 | — | 0.9 | 2M | 200 t/s | 19.2 pts/$ |
| #3 | Gemini 3.2 Pro | Google DeepMind | 96.0 | 1.0 | 0.9 | — | 0.9 | 2M | 200 t/s | 19.2 pts/$ |
| #2 | GPT-5.6 | OpenAI | 96.0 | 1.0 | 0.9 | 0.9 | 0.9 | 2M | 145 t/s | 12.8 pts/$ |
| #2 | GPT-5.6 | OpenAI | 96.0 | 1.0 | 0.9 | 0.9 | 0.9 | 2M | 145 t/s | 12.8 pts/$ |
| #2 | GPT-5.6 | OpenAI | 96.0 | 1.0 | 0.9 | 0.9 | 0.9 | 2M | 145 t/s | 12.8 pts/$ |
| #1 | Gemini 3.2 Pro | Google DeepMind | 96.0 | 1.0 | 0.9 | — | 0.9 | 2M | 200 t/s | 19.2 pts/$ |