Which AI should you use?

Start with what you want to get done. We'll match your task to the evidence, not to the biggest marketing budget.

In a hurry? Just tell me: two quick choices, one answer.

What do you need help with?

Here are the options worth considering.

We've ranked these using the tests that are relevant to your task. Each test has limits, so treat this as a useful starting point, not a guarantee.

In short

For research and checking facts, our pick is Gemini 4 Argon. GPT-6.1 Sol and Claude Opus 5.5 are too close to call against it. If you would rather use a model with open weights, DeepSeek-V4-Pro (High, 20260813) is the best of those we can see. We have verified prices for Gemini 4 Argon and GPT-6.1 Sol only; for the rest we cannot say what they cost.

For Research and checking facts, we looked at short factual questions first.

Best-supported choice Cheapest verified option

Gemini 4 Argon by Google DeepMind

Scored highest on short factual questions among the models with enough evidence, and none of its other relevant results put it behind the leaders.

Worth knowing: Not enough data for fixing real software bugs, so part of the picture is missing.

Price:
  • Google AI Pro: GBP 18.99 per month (price source, verified 28 Oct 2026)
  • Google AI Ultra (from): GBP 79.99 per month (price source, verified 28 Oct 2026)
How we know: scores and sources
Test results for Gemini 4 Argon relevant to research and checking facts
TestResultRatingSetting usedSource
SimpleQA Verified76.0% share of questions correctTop tierhighSource for Gemini 4 Argon SimpleQA Verified score (epoch)
GPQA Diamond96.4% share of questions correctTop tierhighSource for Gemini 4 Argon GPQA Diamond score (epoch)
Arena text: overall preference1,529 Arena scoreTop tierHighSource for Gemini 4 Argon Arena text: overall preference score (lmarena)

See everything about Gemini 4 Argon

Also worth a look

Good-enough alternative Close call with our top pick

GPT-6.1 Sol by OpenAI

Rated Top tier on short factual questions.

Worth knowing: On everyday chat votes it sits in the middle of the listed models.

Price:
  • ChatGPT Plus: GBP 20.00 per month (price source, verified 29 Oct 2026)
  • ChatGPT Pro (from): GBP 89.00 per month (price source, verified 29 Oct 2026)
How we know: scores and sources
Test results for GPT-6.1 Sol relevant to research and checking facts
TestResultRatingSetting usedSource
SimpleQA Verified73.9% share of questions correctTop tiermaxSource for GPT-6.1 Sol SimpleQA Verified score (epoch)
GPQA Diamond95.4% share of questions correctTop tiermaxSource for GPT-6.1 Sol GPQA Diamond score (epoch)
Arena text: overall preference1,446 Arena scoreMiddleMaxSource for GPT-6.1 Sol Arena text: overall preference score (lmarena)

Full profile for GPT-6.1 Sol

Good-enough alternative Close call with our top pick

Claude Opus 5.5 by Anthropic

Rated Top tier on short factual questions.

Worth knowing: On graduate-level science questions it sits in the middle of the listed models.

Price:

Price not independently verified

How we know: scores and sources
Test results for Claude Opus 5.5 relevant to research and checking facts
TestResultRatingSetting usedSource
SimpleQA Verified72.2% share of questions correctTop tiermaxSource for Claude Opus 5.5 SimpleQA Verified score (epoch)
GPQA Diamond90.6% share of questions correctMiddlemaxSource for Claude Opus 5.5 GPQA Diamond score (epoch)
Arena text: overall preference1,516 Arena scoreTop tierHighSource for Claude Opus 5.5 Arena text: overall preference score (lmarena)

Full profile for Claude Opus 5.5

Best open-weights option

DeepSeek-V4-Pro (High, 20260813) by DeepSeek

Rated Strong on short factual questions. Its weights are published under an open licence, so it can be offered by more than one provider. We have not verified any price or free access.

Worth knowing: On everyday chat votes it sits in the middle of the listed models.

Price:

Price not independently verified

How we know: scores and sources
Test results for DeepSeek-V4-Pro (High, 20260813) relevant to research and checking facts
TestResultRatingSetting usedSource
SimpleQA Verified52.9% share of questions correctStrongmaxSource for DeepSeek-V4-Pro (High, 20260813) SimpleQA Verified score (epoch)
GPQA Diamond91.7% share of questions correctStrongmaxSource for DeepSeek-V4-Pro (High, 20260813) GPQA Diamond score (epoch)
Arena text: overall preference1,446 Arena scoreMiddleHigh, 20260813Source for DeepSeek-V4-Pro (High, 20260813) Arena text: overall preference score (lmarena)

Full profile for DeepSeek-V4-Pro (High, 20260813)

The “Cheapest verified option” label means the lowest verified monthly plan price among the options shown that have a verified price. Options without a verified price are not compared, so they may cost less. Check each plan's own limits before deciding.

Other models with enough evidence (39)

These have a result for the main test but were not shown above. They rated Middle or Behind the leaders on it, or fell outside the top few. Listed in score order.

Models we could not rank for this task (34)

Missing evidence is not a low score. These models are left out of the ranking because the tests for this task have no result for them.

How this was decided: our recommendation rules. Results are relative to the 77 models listed here.

Your choice stays in your browser. Nothing you select here is sent anywhere.