Claude 3 Haiku benchmarks, pricing and specs
Claude 3 Haiku is an Anthropic model released on March 13, 2024 that takes text and image input and produces text. It is now listed as a legacy model. It costs $0.25 per 1M input tokens and $1.25 per 1M output tokens and accepts up to 200,000 tokens of context. It scores 1262 Elo on Arena Text (Arena, measured October 8, 2026).
- Per 1M input tokens
- $0.25
- Per 1M output tokens
- $1.25
- Context window
- 200K
- Arena Text
- 1262 Elo
Data verified October 12, 2026 · Dataset 2026-10-12
Key facts
- Provider
- Anthropic
- Model family
- Claude 3
- Release date
- March 13, 2024
- Status
- Legacy
- Access
- API only
- API model id
- Not reported
- Context window
- 200,000 tokens
- Max output
- 4,096 tokens
- Input
- text, image
- Output
- text
- Task families
- text, multimodal
- Declared capabilities
- reasoning
Claude 3 Haiku pricing
It is the #5 cheapest of 38 current text models priced per 1M input tokens.
Independent benchmark results
Scores come from independent evaluators that ran the model themselves, pinned to a dated snapshot. Rank is among current models in the same snapshot. Anthropic's own launch figures are listed under strengths with the publisher named, because they are not measured the same way.
| Benchmark | Score | Rank | Measured | Source |
|---|---|---|---|---|
| Arena Text | 1262 Elo | Not ranked | October 8, 2026 | Arena · 2026-10-12 |
| GPQA Diamond | 36.3% | Not ranked | October 12, 2026 | Epoch AI · 2026-10-12 |
Strengths, cautions and best uses
Strengths
- Claude 3 Haiku records API access for text and image inputs and a published 200,000-token context window
Cautions
- This record is marked legacy; confirm current availability and migration requirements before adoption
Best for
- Historical cost-sensitive text-and-image API comparison
Alternatives to Claude 3 Haiku
The closest current textmodels by price. The cheapest model that scores as well or better is listed first; failing that, the cheapest one within 3% of the benchmark's scale.
Cheaper, same or higher score
GPT-4.1 nano
- 60% cheaper per 1M input tokens, $0.10 vs $0.25
- 60 points higher on Arena Text, 1322 vs 1262
- 1.05M context window vs 200K
- From OpenAI
Gemini 3.1 Flash-Lite
- Same price per 1M input tokens, $0.25
- 45.5 points higher on GPQA Diamond, 81.8 vs 36.3
- 1.05M context window vs 200K
- From Google
DeepSeek-V4.1-Flash
- 20% more per 1M input tokens, $0.30 vs $0.25
- 213 points higher on Arena Text, 1475 vs 1262
- 1M context window vs 200K
- From DeepSeek
Gemini 3.5 Flash-Lite
- 20% more per 1M input tokens, $0.30 vs $0.25
- 193 points higher on Arena Text, 1455 vs 1262
- 1.05M context window vs 200K
- From Google
Other Claude 3 models
Claude 3 Haiku FAQ
How much does Claude 3 Haiku cost?
Claude 3 Haiku costs $0.25 per 1M input tokens and $1.25 per 1M output tokens at the standard tier. It is the #5 cheapest of 38 current text models priced per 1M input tokens. Source: Anthropic, verified August 6, 2026.
What is the context window of Claude 3 Haiku?
Claude 3 Haiku accepts up to 200,000 tokens of context. It can return up to 4,096 tokens in one response. That ranks #40 of 45 current text models by context length.
How does Claude 3 Haiku score on independent benchmarks?
Claude 3 Haiku scores 1262 Elo on Arena Text (Arena, measured October 8, 2026) and 36.3% on GPQA Diamond (Epoch AI, measured October 12, 2026). These are independent evaluations, not Anthropic's own launch figures.
Is Claude 3 Haiku open source or API only?
Claude 3 Haiku is available through Anthropic's API only; its weights are not published.
What are the best alternatives to Claude 3 Haiku?
The closest current alternatives by price and measured performance are GPT-4.1 nano (60% cheaper per 1M input tokens, $0.10 vs $0.25), Gemini 3.1 Flash-Lite (same price per 1M input tokens, $0.25), DeepSeek-V4.1-Flash (20% more per 1M input tokens, $0.30 vs $0.25) and Gemini 3.5 Flash-Lite (20% more per 1M input tokens, $0.30 vs $0.25).
Sources
- Claude 3 Haiku · Anthropic · official · verified August 6, 2026
- Claude models overview · Anthropic · official · verified August 6, 2026
- Claude pricing · Anthropic · official · verified August 6, 2026
- Epoch AI Benchmarking Hub data (benchmark_data.zip) · Epoch AI · independent · verified October 12, 2026 · CC-BY-4.0
- Arena leaderboard dataset (lmarena-ai/leaderboard-dataset) · Arena · independent · verified October 12, 2026 · CC-BY-4.0
Cite this page
Van Data Team. "Claude 3 Haiku Benchmarks, Pricing and Specs." AI Model Benchmark Dataset 2026-10-12. https://vandatateam.com/ai-model-benchmarks/claude-3-haiku
Reuse the data: JSON · CSV · or explore every model in the AI model benchmark comparison.
Production evaluation
Deciding whether Claude 3 Haiku fits your workload?
Public benchmarks narrow the shortlist. We test the finalists on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- Evaluation cases and acceptance thresholds tied to real work
- Model routing, fallbacks and cost controls