Cheapest AI to analyze a 200-row spreadsheet (Oct 2026)
The cheapest model overall is Summarize at 0.03¢ per task. If you want a top-quartile model, the cheapest is DeepSeek V4.1 Flash at 0.26¢.
10 cheapest models
Any model with a price, including small and older ones.
| # | Model | Cost per task | Text score | Tokens in / out | Basis | |
|---|---|---|---|---|---|---|
| 1 | S SummarizeScaledown | ≥0.03¢$0.25 / 1,000 | - | 6,357 / 16,000 | estimated capped | |
| 2 | S ExtractScaledown | ≥0.03¢$0.25 / 1,000 | - | 6,357 / 16,000 | estimated capped | |
| 3 |  Mistral NemoMistral | 0.05¢$0.46 / 1,000 | - | 4,774 / 10,700 | estimated | |
| 4 |  gpt-oss-20bOpenAI | 0.08¢$0.76 / 1,000 | 1317 | 4,779 / 7,472 | estimated best value | |
| 5 | S Llama 3 8B LunarisSao10K | ≥0.11¢$1.05 / 1,000 | - | 6,357 / 16,000 | estimated capped | |
| 6 | Mistral Small 3Mistral | 0.11¢$1.09 / 1,000 | 1274 | 4,774 / 10,700 | estimated | |
| 7 |  Gemma 3 4BGoogle | 0.11¢$1.13 / 1,000 | 1303 | 6,304 / 8,118 | estimated | |
| 8 | I Ling 3.0 FlashInclusionai | ≥0.11¢$1.14 / 1,000 | - | 6,350 / 16,000 | estimated capped | |
| 9 | Gemma 3 12BGoogle | 0.15¢$1.53 / 1,000 | 1342 | 6,304 / 8,118 | estimated best value | |
| 10 | Ministral 3 3B 2512Mistral | 0.15¢$1.55 / 1,000 | - | 4,774 / 10,700 | estimated |
Cheapest high-quality models
Only models in the top 25% of LMArena Text scores (≥ 1458).
| # | Model | Cost per task | Text score | Tokens in / out | Basis | |
|---|---|---|---|---|---|---|
| 1 |  DeepSeek V4.1 FlashDeepSeek | ≥0.26¢$2.65 / 1,000 | 1475 | 4,801 / 16,000 | measured best value capped | |
| 2 |  GLM 5.3 FlashZ.ai | ≥0.34¢$3.44 / 1,000 | 1474 | 4,900 / 16,000 | measured capped | |
| 3 | DeepSeek V4 Pro 0423DeepSeek | ≥0.77¢$7.68 / 1,000 | 1458 | 4,801 / 16,000 | estimated capped | |
| 4 |  MiMo-V2.6-ProXiaomi | ≥1.6¢$16.09 / 1,000 | 1479 | 4,984 / 16,001 | estimated best value capped | |
| 5 | MiMo-V2.5-ProXiaomi | ≥1.6¢$16.09 / 1,000 | 1467 | 4,984 / 16,001 | estimated capped | |
| 6 | Gemini 3 Flash PreviewGoogle | 2.8¢$27.51 / 1,000 | 1473 | 6,304 / 8,118 | estimated | |
| 7 | Gemini 3.7 FlashGoogle | 3.5¢$35.17 / 1,000 | 1488 | 6,304 / 8,118 | estimated best value | |
| 8 | Gemini 3.6 FlashGoogle | 3.5¢$35.17 / 1,000 | 1483 | 6,304 / 8,118 | estimated | |
| 9 | GPT-6 SolOpenAI | 3.7¢$37.50 / 1,000 | 1459 | 4,779 / 2,794 | estimated | |
| 10 | Gemini 3.8 FlashGoogle | 3.9¢$38.69 / 1,000 | 1496 | 6,304 / 8,118 | measured best value |
Cheapest model of each lab
| Model | Cost per task | Score |
|---|---|---|
Mistral NemoMistral | 0.05¢$0.46 / 1,000 | - |
gpt-oss-20bOpenAI | 0.08¢$0.76 / 1,000 | 1317 |
Gemma 3 4BGoogle | 0.11¢$1.13 / 1,000 | 1303 |
 Llama 3.1 8B InstructMeta | ≥0.16¢$1.60 / 1,000 | 1211 |
 Qwen3.7 FlashAlibaba Qwen | ≥0.23¢$2.27 / 1,000 | - |
DeepSeek V4.1 FlashDeepSeek | ≥0.26¢$2.65 / 1,000 | 1475 |
GLM 5.3 FlashZ.ai | ≥0.34¢$3.44 / 1,000 | 1474 |
MiMo-V2.6-FlashXiaomi | ≥0.54¢$5.36 / 1,000 | 1451 |
 Claude Haiku 5.5Anthropic | ≥0.85¢$8.54 / 1,000 | - |
 MiniMax M2.5MiniMax | ≥1.7¢$16.53 / 1,000 | 1391 |
 Kimi K2 ThinkingMoonshot AI | 2.4¢$24.26 / 1,000 | - |
 Grok Build 0.1xAI | 6.9¢$68.59 / 1,000 | - |
FAQ
What is the cheapest AI to analyze a 200-row spreadsheet?
The cheapest model overall is Summarize at 0.03¢ per task. If you want a top-quartile model, the cheapest is DeepSeek V4.1 Flash at 0.26¢.
Is the cheapest model good enough?
Often not for hard tasks. That is why this page also lists the cheapest models in the top quarter of LMArena scores and the cheapest model of every major lab.