Model Comparison

Compare AI models side by side

Pick up to 3 models and compare input cost, output cost, context window, speed, and more. The best value in each row is highlighted. Share your comparison with a link.

	AnthropicPremium Claude Opus 4.6 The current #1 coding model by SWE-bench — use when quality is non-negotiable.	GoogleBudget Gemini 3.1 Flash Best cheap AI for broad day-to-day work — now with 1M context.	AnthropicBudget Claude 4 Haiku Best low-cost writing option for fast-moving content teams.
Input cost / 1M tokens	$15.00/1M	$0.50/1M	$0.80/1M
Output cost / 1M tokens	$75.00/1M	$3.00/1M	$4.00/1M
Context window	1M tokens	1M tokens	200k tokens
Speed	Deliberate	Very fast	Very fast
Price tier	Premium	Budget	Budget
Best for	Agentic coding, complex multi-step reasoning, and deep research	High-volume everyday AI usage where speed and cost both matter	Fast budget writing, support automation, and cost-sensitive Anthropic integrations
Last verified	Mar 23, 2026	Mar 23, 2026	Mar 23, 2026
Verdict	The strongest coding model available by benchmark. Justified for high-stakes engineering work where quality has real financial consequences. For most teams, Sonnet 4.6 at 5× lower cost is the smarter default.	The best all-around budget model for most teams. Faster than its predecessor, cheaper, and with a 1M context window that outclasses every other budget option.	The best pick when you want Anthropic quality at a budget price point — especially for writing-heavy automations.
	View full profile Try Claude Opus 4.6	View full profile Try Gemini 3.1 Flash	View full profile Try Claude 4 Haiku

AnthropicPremium

Claude Opus 4.6

The current #1 coding model by SWE-bench — use when quality is non-negotiable.

Input cost / 1M tokens: $15.00/1M
Output cost / 1M tokens: $75.00/1M
Context window✓ best: 1M tokens
Speed: Deliberate
Price tier: Premium

The strongest coding model available by benchmark. Justified for high-stakes engineering work where quality has real financial consequences. For most teams, Sonnet 4.6 at 5× lower cost is the smarter default.

Full profile Try Claude Opus 4.6

GoogleBudget

Gemini 3.1 Flash

Best cheap AI for broad day-to-day work — now with 1M context.

Input cost / 1M tokens✓ best: $0.50/1M
Output cost / 1M tokens✓ best: $3.00/1M
Context window✓ best: 1M tokens
Speed✓ best: Very fast
Price tier✓ best: Budget

The best all-around budget model for most teams. Faster than its predecessor, cheaper, and with a 1M context window that outclasses every other budget option.

Full profile Try Gemini 3.1 Flash

AnthropicBudget

Claude 4 Haiku

Best low-cost writing option for fast-moving content teams.

Input cost / 1M tokens: $0.80/1M
Output cost / 1M tokens: $4.00/1M
Context window: 200k tokens
Speed✓ best: Very fast
Price tier✓ best: Budget

The best pick when you want Anthropic quality at a budget price point — especially for writing-heavy automations.

Full profile Try Claude 4 Haiku

Share this comparison

What to compare next

Top coding models Budget vs balanced The three big defaults Budget picks

	AnthropicPremium Claude Opus 4.6 The current #1 coding model by SWE-bench — use when quality is non-negotiable.	GoogleBudget Gemini 3.1 Flash Best cheap AI for broad day-to-day work — now with 1M context.	AnthropicBudget Claude 4 Haiku Best low-cost writing option for fast-moving content teams.
Input cost / 1M tokens	$15.00/1M	$0.50/1M	$0.80/1M
Output cost / 1M tokens	$75.00/1M	$3.00/1M	$4.00/1M
Context window	1M tokens	1M tokens	200k tokens
Speed	Deliberate	Very fast	Very fast
Price tier	Premium	Budget	Budget
Best for	Agentic coding, complex multi-step reasoning, and deep research	High-volume everyday AI usage where speed and cost both matter	Fast budget writing, support automation, and cost-sensitive Anthropic integrations
Last verified	Mar 23, 2026	Mar 23, 2026	Mar 23, 2026
Verdict	The strongest coding model available by benchmark. Justified for high-stakes engineering work where quality has real financial consequences. For most teams, Sonnet 4.6 at 5× lower cost is the smarter default.	The best all-around budget model for most teams. Faster than its predecessor, cheaper, and with a 1M context window that outclasses every other budget option.	The best pick when you want Anthropic quality at a budget price point — especially for writing-heavy automations.
	View full profile Try Claude Opus 4.6	View full profile Try Gemini 3.1 Flash	View full profile Try Claude 4 Haiku