Type ranking
Best Non-Reasoning LLMs in October 2026
As of October 2026, Grok 4.20 (Non-Reasoning) leads the Noometry Index with a score of 48.6, ahead of Qwen3 Max at 43.7.
Last verified
Models that answer directly without a separate thinking phase. They are usually faster and cheaper per task.
- Grok 4.20 (Non-Reasoning) 48.6
- Qwen3 Max 43.7
- Qwen3-Next 80B-A3B Instruct 43.0
- GPT-5.3 Chat 42.8
- Grok 4.1 Fast 41.4
- DeepSeek-V3 39.5
- Mistral Large 3 39.1
- Sonar 38.5
- Nova Premier 1.0 38.3
- Qwen3-Coder 480B-A35B Instruct 38.1
Full ranking
51 models
| 1 | xAI Grok 4.20 (Non-Reasoning) | xAI | Proprietary | 1M | 48.6 | $1.25 | $2.50 | 31.1 | 61 |
| 2 |  Alibaba (Qwen) Qwen3 Max | Alibaba (Qwen) | Proprietary | 262K | 43.7 | $1.20 | $6 | 18.2 | 48 |
| 3 | Alibaba (Qwen) Qwen3-Next 80B-A3B Instruct | Alibaba (Qwen) | Open | 131K | 43.0 | $0.50 | $2 | 49.2 | 111 |
| 4 | OpenAI GPT-5.3 Chat | OpenAI | Proprietary | 128K | 42.8 | $1.75 | $14 | 8.88 | — |
| 5 | xAI Grok 4.1 Fast | xAI | Proprietary | 128K | 41.4 | $0.20 | $0.50 | 150 | — |
| 6 |  DeepSeek DeepSeek-V3 | DeepSeek | Open | 164K | 39.5 | $0.24 | $0.90 | 97.5 | 73 |
| 7 |  Mistral AI Mistral Large 3 | Mistral AI | Open | 262K | 39.1 | $0.25 | $0.75 | 104 | 7 |
| 8 |  Perplexity Sonar | Perplexity | Proprietary | 128K | 38.5 | $1 | $1 | 38.5 | — |
| 9 | Amazon Nova Premier 1.0 | Amazon | Proprietary | 1M | 38.3 | $2.50 | $12.50 | 7.67 | 9 |
| 10 | Alibaba (Qwen) Qwen3-Coder 480B-A35B Instruct | Alibaba (Qwen) | Open | 262K | 38.1 | $1.50 | $7.50 | 12.7 | 67 |
| 11 |  Cohere Command A | Cohere | Open | 256K | 36.5 | $2.50 | $10 | 8.33 | 28 |
| 12 | OpenAI GPT-4.1 | OpenAI | Proprietary | 1.05M | 35.9 | $2 | $8 | 10.3 | 116 |
| 13 | Cohere C4ai Aya Expanse 32b | Cohere | Open | 128K | 35.9 | — | — | — | — |
| 14 | Alibaba (Qwen) Qwen Max | Alibaba (Qwen) | Proprietary | 33K | 34.7 | $1.60 | $6.40 | 12.4 | — |
| 15 | Alibaba (Qwen) Qwen3 Coder Next | Alibaba (Qwen) | Open | 262K | 34.3 | $0.12 | $0.80 | 118 | — |
| 16 | Mistral AI Devstral Small 2505 | Mistral AI | Open | 128K | 34.3 | $0.10 | $0.30 | 229 | 88 |
| 17 | OpenAI GPT-4.1 mini | OpenAI | Proprietary | 1.05M | 33.6 | $0.40 | $1.60 | 48.1 | 86 |
| 18 | Cohere Command R+ | Cohere | Open | 128K | 32.4 | $2.50 | $10 | 7.41 | — |
| 19 | Mistral AI Pixtral Large | Mistral AI | Open | 128K | 32.2 | $2 | $6 | 10.7 | — |
| 20 |  Google Gemma 3 12B | Open | 131K | 32.1 | $0.05 | $0.15 | 428 | — | |
| 21 | Mistral AI Mistral Large | Mistral AI | Open | 131K | 31.9 | $2 | $6 | 10.6 | — |
| 22 | Amazon Amazon Nova Lite | Amazon | Proprietary | 300K | 31.9 | $0.06 | $0.24 | 304 | — |
| 23 | Alibaba (Qwen) Qwen2.5 72B Instruct | Alibaba (Qwen) | Open | 131K | 31.9 | $1.40 | $5.60 | 13 | — |
| 24 | Cohere Command R | Cohere | Open | 128K | 31.4 | $0.15 | $0.60 | 120 | — |
| 25 |  Microsoft Phi-4 | Microsoft | Open | 128K | 31.2 | $0.07 | $0.14 | 357 | — |
| 26 | Amazon Amazon Nova Pro | Amazon | Proprietary | 300K | 31.0 | $0.80 | $3.20 | 22.1 | — |
| 27 |  Meta Llama 4 Maverick | Meta | Open | 128K | 30.9 | $0.19 | $0.65 | 102 | 456 |
| 28 | Microsoft Phi-4 Mini | Microsoft | Open | 128K | 30.9 | $0.075 | $0.30 | 235 | — |
| 29 | Google Gemma 3 27B | Open | 131K | 30.8 | $0.08 | $0.16 | 308 | 62 | |
| 30 | Mistral AI Codestral | Mistral AI | Proprietary | 256K | 30.6 | $0.30 | $0.90 | 68 | 271 |
| 31 | Meta Llama-3.3-70B-Instruct | Meta | Open | 128K | 30.6 | $0.10 | $0.32 | 197 | — |
| 32 | OpenAI GPT-4 Turbo | OpenAI | Proprietary | 128K | 30.5 | $10 | $30 | 2.03 | — |
| 33 | Amazon Amazon Nova Micro | Amazon | Proprietary | 128K | 30.4 | $0.035 | $0.14 | 496 | — |
| 34 | Alibaba (Qwen) Qwen2.5 32B Instruct | Alibaba (Qwen) | Open | 131K | 30.1 | $0.70 | $2.80 | 24.6 | — |
| 35 | Alibaba (Qwen) Qwen2.5-VL 72B Instruct | Alibaba (Qwen) | Open | 131K | 29.9 | $2.80 | $8.40 | 7.13 | 43 |
| 36 | Meta Llama 3.1-70B | Meta | Open | 128K | 29.6 | $0.40 | $0.40 | 74.1 | — |
| 37 | OpenAI GPT-4 | OpenAI | Proprietary | 8K | 29.1 | $30 | $60 | 0.78 | — |
| 38 | Alibaba (Qwen) Qwen2.5 7B Instruct | Alibaba (Qwen) | Open | 131K | 29.0 | $0.17 | $0.70 | 94.6 | — |
| 39 | OpenAI GPT-4o | OpenAI | Proprietary | 128K | 28.6 | $2.50 | $10 | 6.53 | — |
| 40 | Mistral AI Ministral 8B | Mistral AI | Open | 262K | 28.2 | $0.15 | $0.15 | 188 | — |
| 41 | Google Gemma 3 4B | Open | 131K | 28.1 | $0.04 | $0.08 | 563 | 72 | |
| 42 | OpenAI GPT-4.1 nano | OpenAI | Proprietary | 1.05M | 27.9 | $0.10 | $0.40 | 159 | 135 |
| 43 | Meta Llama 4 Scout | Meta | Open | 128K | 27.7 | $0.10 | $0.30 | 184 | 272 |
| 44 | Mistral AI Mixtral 8x22B | Mistral AI | Open | 64K | 27.1 | $2 | $6 | 9.05 | — |
| 45 | Mistral AI Mixtral 8x7B | Mistral AI | Open | 32K | 27.1 | $0.70 | $0.70 | 38.8 | — |
| 46 | Mistral AI Mistral Nemo | Mistral AI | Open | 128K | 26.4 | $0.15 | $0.15 | 176 | — |
| 47 | Mistral AI Ministral 3B | Mistral AI | Open | 131K | 26.2 | $0.10 | $0.10 | 262 | — |
| 48 | OpenAI GPT-4o mini | OpenAI | Proprietary | 128K | 25.5 | $0.15 | $0.60 | 97.2 | 120 |
| 49 | OpenAI GPT-3.5-turbo | OpenAI | Proprietary | 16K | 23.2 | $0.50 | $1.50 | 31 | — |
| 50 | Mistral AI Mistral 7B | Mistral AI | Open | 8K | 23.0 | $0.25 | $0.25 | 92.2 | — |
| 51 | Meta Llama 3.1-8B | Meta | Open | 128K | 23.0 | $0.05 | $0.08 | 400 | — |
Sponsored placements are available on pages like this one. Advertise on Noometry
Compare the leaders
Frequently asked questions
What is the best fast, non-thinking model?
As of October 2026, Grok 4.20 (Non-Reasoning) leads the Noometry Index with a score of 48.6, ahead of Qwen3 Max at 43.7.
What are the top 5 in this ranking?
Grok 4.20 (Non-Reasoning), Qwen3 Max, Qwen3-Next 80B-A3B Instruct, GPT-5.3 Chat, Grok 4.1 Fast, in that order, as of October 2026.
How is this list ranked?
By the Noometry Index, which is fitted from every public benchmark result and adjusted for how hard each benchmark is. Models need several independent results to be ranked.