Best LLM
What is the best LLM for knowledge questions?
Claude Fable 5.1, Gemini 4 Argon and GPT-6 Astra are practically level at the top in knowledge questions.
Among the ten best, Gemini 3.8 Flash is the cheapest at $1.50 per million tokens.
The fastest among them is Gemini 3.5 Flash, at 142 tokens per second.
- Ranked models
- 33
- Tests
- 3
- Data as of
- Nov 10, 2026
Ranking: Knowledge
Place and value in the area, plus overall index, price and speed. Higher values are better.
| Rank | Model | Value | Overall index | Price (3:1) | Speed |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.1Top group | 130.1 | 124.1 | $20 | 59 |
| 2 | Gemini 4 ArgonTop group | 127.0 | 121.4 | $4 | 46 |
| 3 | GPT-6 AstraTop group | 126.0 | 120.2 | $20 | 58 |
| 4 | GPT-6.1 SolUpper group | 123.8 | 118.0 | $4 | 44 |
| 5 | Claude Opus 5Upper group | 121.8 | 117.4 | $10 | 71 |
| 6 | Gemini 3.8 FlashMiddle group | 113.3 | 106.8 | $1.50 | 81 |
| 7 | Gemini 3.7 FlashMiddle group | 112.9 | 107.9 | n/a | 72 |
| 8 | GPT-6 SolMiddle group | 109.0 | 109.1 | $4 | 82 |
| 9 | Gemini 3.5 FlashMiddle group | 108.0 | 97.8 | n/a | 142 |
| 10 | Gemini 3.6 FlashMiddle group | 107.8 | 97.6 | $1.50 | 110 |
| 11 | GPT-5.5Middle group | 106.0 | 101.5 | $11.25 | 87 |
| 12 | Grok 4.7Middle group | 105.7 | 103.3 | $3 | 47 |
| 13 | Grok 4.6Middle group | 100.2 | 102.1 | $3 | 58 |
| 14 | Qwen 3.8 Max 0902Middle group | 97.7 | 100.8 | $2.48 | 36 |
| 15 | Grok 4.5Middle group | 96.1 | 94.8 | $3 | 52 |
| 16 | Kimi K3Middle group | 95.4 | 101.1 | $6 | 61 |
| 17 | GLM-5.3Middle group | 90.6 | 100.6 | $2.15 | 81 |
| 18 | DeepSeek-V4-ProMiddle group | 89.4 | 87.2 | $0.99 | 44 |
| 19 | GPT-6 LunaMiddle group | 88.1 | 95.8 | $0.20 | 77 |
| 20 | GPT-5.6 TerraMiddle group | 88.1 | 100.8 | $4.50 | 72 |
Show all 33 ranked models
| Rank | Model | Value | Overall index | Price (3:1) | Speed |
|---|---|---|---|---|---|
| 21 | GLM-5.3-FlashMiddle group | 87.1 | 94.3 | $0.24 | 101 |
| 22 | Claude Sonnet 5Lower group | 84.6 | 92.1 | $4 | 77 |
| 23 | GLM-5.2Lower group | 83.5 | 89.0 | $2.15 | 73 |
| 24 | Claude Haiku 5.5Lower group | 80.9 | 102.8 | $0.20 | 164 |
| 25 | GPT-5.6 LunaLower group | 80.4 | 93.8 | $0.45 | 123 |
| 26 | Gemini 3.5 Flash-LiteLower group | 78.5 | 73.7 | $0.85 | 145 |
| 27 | Qwen 3.8 27BLower group | 77.0 | 82.8 | $0.66 | 66 |
| 28 | MiniMax M3Lower group | 76.7 | 83.0 | $0.52 | 82 |
| 29 | InklingLower group | 75.8 | 72.1 | $1.76 | n/a |
| 30 | Nemotron 3 UltraLower group | 73.8 | 66.8 | $0.93 | 92 |
| 31 | Inkling SmallLower group | 69.7 | 68.7 | $0.52 | 61 |
| 32 | Mistral Medium 3.5Lower group | 36.4 | 46.3 | $3 | 118 |
| 33 | GPT OSS 120BLower group | 35.9 | 43.4 | $0.20 | 92 |
Price in dollars per million tokens, mixed from 3 parts input and 1 part output. Speed in tokens per second.
Claude Fable 5.1Rank 1130.1
Gemini 4 ArgonRank 2127.0
GPT-6 AstraRank 3126.0
GPT-6.1 SolRank 4123.8
Claude Opus 5Rank 5121.8
Gemini 3.8 FlashRank 6113.3
Gemini 3.7 FlashRank 7112.9
GPT-6 SolRank 8109.0
Gemini 3.5 FlashRank 9108.0
Gemini 3.6 FlashRank 10107.8
GPT-5.5Rank 11106.0
Grok 4.7Rank 12105.7
Grok 4.6Rank 13100.2
Qwen 3.8 Max 0902Rank 1497.7
Grok 4.5Rank 1596.1
Kimi K3Rank 1695.4
GLM-5.3Rank 1790.6
DeepSeek-V4-ProRank 1889.4
GPT-6 LunaRank 1988.1
GPT-5.6 TerraRank 2088.1
Which tests count here?
3 tests by independent evaluators feed this area. Each test counts with the result of the model in question.
MMLU Pro
83 models tested · Vals AI
AA-Omniscience
57 models tested · Artificial Analysis
SimpleQA Verified
19 models tested · Epoch AI benchmark run dataset snapshot
Popular comparisons among the best
Other tasks
All LLMs comparedHow the overall index worksCompare several models