Model AI terbaik untuk pengetahuan
Peringkat model bahasa berdasar akurasi dan kedalaman pengetahuan. Diperbarui Agustus 2026. Berdasar data benchmark independen.
| # | Model | Penyedia | Skor | Kecepatan | Harga (output) |
|---|---|---|---|---|---|
| 1 | Claude Mythos Preview | Anthropic | 100.0 | 80 t/s | $15/M |
| 2 | Gemini 3.1 Pro | 97.6 | 136 t/s | $12/M | |
| 3 | Gemini 3 Pro | 92.9 | 120 t/s | $5/M | |
| 4 | Claude Mythos 5 | Anthropic | 75.0 | 50 t/s | $50/M |
| 5 | Qwen3.7 Max | Alibaba | 69.3 | 203 t/s | $7.5/M |
| 6 | Claude Sonnet 4.6 (max) | Anthropic | 67.0 | 50 t/s | $15/M |
| 7 | Qwen3.8 Flash Next | Alibaba | 64.0 | 168 t/s | $0.4/M |
| 8 | GLM-5.3 Flash | Z.ai | 60.5 | 182 t/s | $0.3/M |
| 9 | Claude Fable 5.1 | Anthropic | 58.0 | 85 t/s | $15/M |
| 10 | Gemini 3.8 Flash | 58.0 | 302 t/s | $3.75/M | |
| 11 | Gemini 3.8 Flash (Medium Reasoning) | 58.0 | 315 t/s | $3.75/M | |
| 12 | Gemini 3.8 Flash (Low Reasoning) | 58.0 | 340 t/s | $3.75/M | |
| 13 | Grok 4.7 (high) | xAI | 55.0 | 80 t/s | $15/M |
| 14 | Grok 4.7 | xAI | 55.0 | 95 t/s | $15/M |
| 15 | Granite 4.2 30B Instruct | IBM | 52.0 | 78 t/s | $1.2/M |
| 16 | GPT-5.5 Pro | OpenAI | 50.0 | 60 t/s | $10/M |
| 17 | Qwen3.6 Plus | Alibaba | 48.4 | 55 t/s | $3/M |
| 18 | Agnes 2.5 Pro Beta | Agnes | 48.0 | 65 t/s | $3/M |
| 19 | DeepSeek V4.1 Flash | DeepSeek | 46.0 | 214 t/s | $1.2/M |
| 20 | Claude Fable 5.1 (XHigh) | Anthropic | 45.0 | 78 t/s | $15/M |
Claude Mythos Preview
Gemini 3.1 Pro
Qwen3.7 Max
GLM-5.3 Flash
Grok 4.7 (high)
Granite 4.2 30B Instruct
GPT-5.5 Pro
Agnes 2.5 Pro Beta
DeepSeek V4.1 FlashTop 5 \u2014 Perbandingan cepat
Metodologi
Peringkat berdasar evaluasi benchmark independen. Skor dihitung dari beberapa benchmark terstandar dan dinormalisasi ke 0–100.