Best AI models for data analysis.

Prices checked

For data analysis, GPT-6 Astra is the best measured match of 42 models with published results, from $19 per M tokens at Jiekou, checked .

RankNamePrice per M tokensLinks
1

GPT-6 Astra

Best match

Jiekou · Capability 82 · 1.1M context · 44 providers

$19/M tokens
2

ZenMux · Capability 81 · 1M context · 105 providers

$0.13/M tokens
3

Qwen3.8 27B

Lowest cost

NanoGPT · Capability 75 · 1M context · 76 providers

$0.12/M tokens
4

Claude Fable 5.1

Amazon Bedrock · Capability 83 · 1M context · 42 providers

$20/M tokens
5

Claude Fable 5

Amazon Bedrock · Capability 83 · 1M context · 50 providers

$20/M tokens
6

Muse Spark 1.3

NanoGPT · Capability 82 · 1M context · 21 providers

$0.97/M tokens
7

GPT-5.6 Sol

Azure AI Foundry · Capability 81 · 1.1M context · 52 providers

$8/M tokens
8

GPT-5.5

Poe · Capability 80 · 1.1M context · 53 providers

$10.23/M tokens
9

GPT-5.4

Poe · Capability 78 · 1.1M context · 48 providers

$5.11/M tokens
10

GPT-5.6 Terra

Atlas Cloud · Capability 78 · 1.1M context · 49 providers

$4.50/M tokens

How hard is the task?

Rank again for a quicker or a harder version of it.

How we count

Data work can mean extracting a few fields, writing a query or explaining a pattern in a table. This shortlist emphasizes data analysis, mathematics and reasoning. Choose a model around the decisions it must make, then check the output against the original data.

The list weighs the published results that matter for data analysis and keeps the models close enough to the best one for a balanced task. Price picks the provider after that; it never lifts a weaker model into the list. 382 available models were considered and 42 have comparable evidence.

Each price is the cheapest provider read in the last three days, per M tokens, blended 3 in : 1 out.

Benchmark results do not verify the calculations or conclusions in your output. They also do not guarantee spreadsheet access, code execution or reliable extraction from every document layout.

The ranking uses the same Balanced settings as the Ask page. The Capability methodology has the sources, the exact identity rules, the task weights and the difficulty ranges.

Best match
The strongest published results for this task at a balanced difficulty.
Best value
The cheapest model within five points of the best match.
Lowest cost
The cheapest model in the range this page lists.

Other tasks