Value ranking
Best Value LLMs for Reasoning in October 2026
As of October 2026, Qwen3.7 Flash gives the most reasoning performance per dollar: 28.2 points at a blended $0.05 per million tokens. Nemotron 3 Nano 30B A3B is next.
Last verified
Category score per dollar of blended API price, among models in the top half of the reasoning ranking.
- Qwen3.7 Flash 512.2 pts/$
- Nemotron 3 Nano 30B A3B 310.1 pts/$
- Nemotron 3.5 Lightning 306.6 pts/$
- Granite 4.2 8B 247.6 pts/$
- GPT-6 Luna 241.1 pts/$
- MiMo-V2.6-Flash 208.3 pts/$
- DeepSeek V4 Flash 204.6 pts/$
- GLM-5.3-Flash 201.9 pts/$
- Qwen3.5-Flash 192.8 pts/$
- DeepSeek V4.1 Flash 191.1 pts/$
Full ranking
| # | Model | Score | Input $/M | Output $/M | Points per $ |
|---|---|---|---|---|---|
| 1 | Qwen3.7 Flash | 28.2 | $0.03 | $0.13 | 512.2 |
| 2 | Nemotron 3 Nano 30B A3B | 27.1 | $0.05 | $0.20 | 310.1 |
| 3 | Nemotron 3.5 Lightning | 26.8 | $0.05 | $0.20 | 306.6 |
| 4 | Granite 4.2 8B | 26.6 | $0.06 | $0.25 | 247.6 |
| 5 | GPT-6 Luna | 48.2 | $0.10 | $0.50 | 241.1 |
| 6 | MiMo-V2.6-Flash | 36.5 | $0.14 | $0.28 | 208.3 |
| 7 | DeepSeek V4 Flash | 53.7 | $0.15 | $0.60 | 204.6 |
| 8 | GLM-5.3-Flash | 48.0 | $0.15 | $0.50 | 201.9 |
| 9 | Qwen3.5-Flash | 33.7 | $0.10 | $0.40 | 192.8 |
| 10 | DeepSeek V4.1 Flash | 50.2 | $0.15 | $0.60 | 191.1 |
| 11 | Hy3 | 26.1 | $0.0825 | $0.33 | 180.8 |
| 12 | Gemma 4 31B IT | 27.2 | $0.09 | $0.34 | 178.4 |
| 13 | Claude Haiku 5.5 | 35.2 | $0.10 | $0.50 | 175.9 |
| 14 | MiMo-V2-Omni | 29.7 | $0.14 | $0.28 | 169.9 |
| 15 | MiMo-V2.5 | 28.6 | $0.14 | $0.28 | 163.7 |
| 16 | Grok 4.1 Fast | 43.4 | $0.20 | $0.50 | 157.6 |
| 17 | MiMo-V2-Flash | 24.9 | $0.14 | $0.28 | 142.1 |
| 18 | GPT-5.6 Luna | 47.6 | $0.20 | $1.20 | 105.9 |
| 19 | MiMo-V2.6-Pro | 43.1 | $0.43 | $0.87 | 79.4 |
| 20 | Qwen3.6 Flash | 29.0 | $0.19 | $1.13 | 68.7 |
| 21 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 | 26.8 | $0.40 | $0.40 | 67.0 |
| 22 | DeepSeek-V3.1 | 27.9 | $0.25 | $0.95 | 65.7 |
| 23 | Mercury 2 | 23.8 | $0.25 | $0.75 | 63.5 |
| 24 | GLM-4.6V | 27.6 | $0.30 | $0.90 | 61.4 |
| 25 | Inkling-Small | 38.6 | $0.45 | $1.20 | 60.6 |
| 26 | DeepSeek-V3.1-Terminus | 26.4 | $0.27 | $1 | 58.4 |
| 27 | MiniMax-M3 | 30.1 | $0.30 | $1.20 | 57.3 |
| 28 | DeepSeek V4 Pro | 56.5 | $0.66 | $1.98 | 57.0 |
| 29 | GLM-4.5-Air | 24.1 | $0.20 | $1.10 | 56.6 |
| 30 | Qwen3.7 Plus | 39.3 | $0.40 | $1.60 | 56.1 |
| 31 | Solar Pro4 | 28.5 | $0.30 | $1.20 | 54.2 |
| 32 | GPT-5.4 nano | 23.7 | $0.20 | $1.25 | 51.3 |
| 33 | Gemini 3.8 Flash | 76.9 | $0.75 | $3.75 | 51.2 |
| 34 | Qwen3.6 35B-A3B | 28.0 | $0.25 | $1.49 | 50.2 |
| 35 | MiMo-V2.5-Pro | 26.8 | $0.43 | $0.87 | 49.2 |
| 36 | Qwen Plus | 28.4 | $0.40 | $1.20 | 47.3 |
| 37 | Gemini 3.7 Flash | 70.0 | $0.75 | $3.75 | 46.7 |
| 38 | Gemini 3 Flash Preview | 49.2 | $0.50 | $3 | 43.7 |
| 39 | Gemini 3.6 Flash | 58.8 | $0.75 | $3.75 | 39.2 |
| 40 | DeepSeek-V3.2-Speciale | 32.9 | $0.58 | $1.68 | 38.5 |
| 41 | Qwen3.8 27B | 41.0 | $0.99 | $1.49 | 36.8 |
| 42 | Qwen3.5 Plus | 32.8 | $0.40 | $2.40 | 36.5 |
| 43 | Qwen3.5 35B-A3B | 24.6 | $0.25 | $2 | 35.7 |
| 44 | Qwen3-Next 80B-A3B Instruct | 31.1 | $0.50 | $2 | 35.5 |
| 45 | GPT-5 Mini | 23.9 | $0.25 | $2 | 34.8 |
| 46 | Kimi K2.5 | 31.2 | $0.45 | $2.25 | 34.7 |
| 47 | Nemotron 3 Ultra | 31.0 | $0.50 | $2.20 | 33.5 |
| 48 | Grok 4.20 (Non-Reasoning) | 52.3 | $1.25 | $2.50 | 33.5 |
| 49 | Qwen3.5 27B | 27.5 | $0.30 | $2.40 | 33.4 |
| 50 | Gemini 3.5 Flash Lite | 27.8 | $0.30 | $2.50 | 32.7 |
| 51 | Nova 2 Lite | 27.5 | $0.30 | $2.50 | 32.4 |
| 52 | GLM-4.5V | 27.4 | $0.60 | $1.80 | 30.4 |
| 53 | GLM-4.5 | 28.6 | $0.60 | $2.20 | 28.6 |
| 54 | Hy4 preview | 31.9 | $0.75 | $2.25 | 28.3 |
| 55 | Step 5 Preview | 40.0 | $1 | $2.70 | 28.1 |
| 56 | Grok 4.20 Multi-Agent | 43.9 | $1.25 | $2.50 | 28.1 |
| 57 | MiniMax M1 | 26.9 | $0.55 | $2.20 | 27.9 |
| 58 | Muse Spark 1.3 | 54.0 | $1.25 | $4.25 | 27.0 |
| 59 | Qwen3.6 Plus | 29.3 | $0.50 | $3 | 26.0 |
| 60 | Grok Build 0.1 | 32.2 | $1 | $2 | 25.8 |
| 61 | Muse Spark 1.2 | 51.3 | $1.25 | $4.25 | 25.6 |
| 62 | Qwen3.5 397B-A17B | 34.5 | $0.60 | $3.60 | 25.5 |
| 63 | Qwen3.5 122B-A10B | 27.2 | $0.40 | $3.20 | 24.7 |
| 64 | GLM-4.7 | 24.3 | $0.60 | $2.20 | 24.3 |
| 65 | Qwen3-VL 235B-A22B | 29.3 | $0.70 | $2.80 | 23.9 |
| 66 | GLM-4.6 | 23.7 | $0.60 | $2.20 | 23.7 |
| 67 | Kimi K2.6 | 40.5 | $0.95 | $4 | 23.7 |
| 68 | Muse Spark 1.1 | 47.1 | $1.25 | $4.25 | 23.6 |
| 69 | Grok 4.3 | 35.9 | $1.25 | $2.50 | 23.0 |
| 70 | Kimi K2.7 Code | 39.0 | $0.95 | $4 | 22.8 |
| 71 | GLM-5.3 | 46.1 | $1.40 | $4.40 | 21.4 |
| 72 | Seed 2.0 Pro | 24.1 | $0.50 | $3 | 21.4 |
| 73 | GPT-6.1 Sol | 81.9 | $2 | $10 | 20.5 |
| 74 | Grok 4.6 | 61.4 | $2 | $6 | 20.5 |
| 75 | GLM-5.2 | 42.3 | $1.40 | $4.40 | 19.7 |
| 76 | Grok 4.5 | 56.1 | $2 | $6 | 18.7 |
| 77 | Gemini 3.5 Flash | 62.8 | $1.50 | $9 | 18.6 |
| 78 | Qwen3.6 27B | 25.0 | $0.60 | $3.60 | 18.5 |
| 79 | GPT-6 Sol | 74.0 | $2 | $10 | 18.5 |
| 80 | GLM-5.1 | 39.1 | $1.40 | $4.40 | 18.2 |
| 81 | Qwen3.8 Max | 54.4 | $2 | $6 | 18.1 |
| 82 | GPT-5.4 mini | 30.4 | $0.75 | $4.50 | 18.0 |
| 83 | GLM-5 | 27.6 | $1 | $3.20 | 17.8 |
| 84 | Grok 4.7 | 49.1 | $2 | $6 | 16.4 |
| 85 | Gemini 3.1 Pro Preview | 71.7 | $2 | $12 | 15.9 |
| 86 | Inkling | 40.4 | $1.87 | $4.68 | 15.7 |
| 87 | GLM-5V-Turbo | 29.7 | $1.20 | $4 | 15.6 |
| 88 | Qwen3.6 Max Preview | 41.7 | $1.30 | $7.80 | 14.2 |
| 89 | Claude Sonnet 5.5 | 54.0 | $2 | $10 | 13.5 |
| 90 | GPT-5.6 Terra | 60.7 | $2 | $12 | 13.5 |
| 91 | Qwen3.7 Max | 49.2 | $2.50 | $7.50 | 13.1 |
| 92 | o4-mini | 24.6 | $1.10 | $4.40 | 12.8 |
| 93 | Claude Sonnet 5 | 49.1 | $2 | $10 | 12.3 |
| 94 | GPT-5.1 | 39.8 | $1.25 | $10 | 11.6 |
| 95 | GPT-5 | 38.3 | $1.25 | $10 | 11.1 |
| 96 | GPT-5.4 | 61.8 | $2.50 | $15 | 11.0 |
| 97 | Kimi K3 | 63.0 | $3 | $15 | 10.5 |
| 98 | GPT-5.2 | 50.2 | $1.75 | $14 | 10.4 |
| 99 | Claude Opus 5.5 | 80.2 | $4 | $20 | 10.0 |
| 100 | GPT-5.6 Sol | 74.8 | $4 | $20 | 9.3 |
| 101 | o3 | 32.0 | $2 | $8 | 9.1 |
| 102 | GPT-5-Codex | 30.9 | $1.25 | $10 | 9.0 |
| 103 | Qwen Max | 25.1 | $1.60 | $6.40 | 9.0 |
| 104 | Qwen3-Coder 480B-A35B Instruct | 25.5 | $1.50 | $7.50 | 8.5 |
| 105 | Gemini 2.5 Pro | 28.8 | $1.25 | $10 | 8.4 |
| 106 | Mistral Medium | 24.0 | $1.50 | $7.50 | 8.0 |
| 107 | Claude Opus 5 | 77.2 | $5 | $25 | 7.7 |
| 108 | Claude Sonnet 4.6 | 46.1 | $3 | $15 | 7.7 |
| 109 | Claude Opus 4.8 | 64.7 | $5 | $25 | 6.5 |
| 110 | GPT-5.5 | 72.8 | $5 | $30 | 6.5 |
| 111 | GPT-5.3 Chat | 28.5 | $1.75 | $14 | 5.9 |
| 112 | Claude Opus 4.6 | 57.8 | $5 | $25 | 5.8 |
| 113 | Claude Opus 4.7 | 53.8 | $5 | $25 | 5.4 |
| 114 | Claude Sonnet 4.5 | 26.9 | $3 | $15 | 4.5 |
| 115 | Claude Opus 4.5 | 42.6 | $5 | $25 | 4.3 |
| 116 | GPT-6 Astra | 85.1 | $10 | $50 | 4.3 |
| 117 | Claude Fable 5 | 76.8 | $10 | $50 | 3.8 |
| 118 | Claude Fable 5.1 | 76.7 | $10 | $50 | 3.8 |
| 119 | GPT-5.5 Pro | 73.3 | $30 | $180 | 1.1 |
| 120 | Claude Opus 4.1 | 32.2 | $15 | $75 | 1.1 |
| 121 | o1 | 27.9 | $15 | $60 | 1.1 |
| 122 | GPT-5.4 Pro | 70.7 | $30 | $180 | 1.0 |
| 123 | GPT-5 Pro | 38.9 | $15 | $120 | 0.9 |
| 124 | Claude Opus 4 | 27.3 | $15 | $75 | 0.9 |
| 125 | GPT-5.2 Pro | 51.5 | $21 | $168 | 0.9 |
| 126 | o3-pro | 23.8 | $20 | $80 | 0.7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Compare the leaders
Frequently asked questions
What is the most cost-effective AI model for reasoning?
As of October 2026, Qwen3.7 Flash gives the most reasoning performance per dollar: 28.2 points at a blended $0.05 per million tokens. Nemotron 3 Nano 30B A3B is next.
What are the top 5 in this ranking?
Qwen3.7 Flash, Nemotron 3 Nano 30B A3B, Nemotron 3.5 Lightning, Granite 4.2 8B, GPT-6 Luna, in that order, as of October 2026.
How is this list ranked?
By score per dollar: the model’s score divided by its blended API price (three parts input to one part output), among models in the top half by score.