claude-3-haiku-20240307: benchmark scores, price and ranking

claude-3-haiku-20240307 from Anthropic ranks #736 of 743 LLMs on the LLMs Tiger aggregate benchmark score (39.9), based on 6 benchmark results captured up to 2026-10-13.

Organisation: Anthropic
Type: Fast / low-cost (flash) model
Open weights: No
Released: 2024-03-13 (source: curated)
Cheapest API price: $0.250 in / $1.25 out per million tokens (via vertex_ai, observed 2026-10-13T20:47Z)
Aggregate rank: #736 of 743 (score 39.9)

Relative to other models, claude-3-haiku-20240307 is strongest on Aider Code Editing (#64 of 92) and weakest on WeirdML (#104 of 109).

Across the 6 benchmarks where at least ten models are scored, claude-3-haiku-20240307 is in the top 10% on 0 and in the top half on 0.

Its cheapest listed input price, $0.250 per million tokens, is lower than that of 64% of the 427 models with a listed price.

claude-3-haiku-20240307 benchmark results

claude-3-haiku-20240307 benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
Aider Code Editing	47.4%	#64 of 92	Aider Aider Code Editing leaderboard	2026-10-13
GPQA Diamond	36.3%	#232 of 275	Epoch AI Benchmarking Hub	2026-10-13
DTBench	50.1%	#130 of 146	Epoch AI Benchmarking Hub	2026-10-13
LMCA	8.8%	#98 of 108	Epoch AI Benchmarking Hub	2026-10-13
OTIS Mock AIME	1.8%	#176 of 189	Epoch AI Benchmarking Hub	2026-10-13
WeirdML	9.8%	#104 of 109	Epoch AI Benchmarking Hub	2026-10-13

About claude-3-haiku-20240307

claude-3-haiku-20240307 is a smaller, lower-cost model from Anthropic. It was released on 2024-03-13, according to the developer's announcement. The cheapest API listing we track charges $0.250 per million input tokens and $1.25 per million output tokens, through vertex_ai. We hold 6 benchmark results for it across 4 areas, and none is in the top tenth of its board.

Where claude-3-haiku-20240307 stands

On our aggregate score, claude-3-haiku-20240307 is #736 of 743 ranked models, in the lower half. Its score is 39.9; the scale is centred on 50, so it sits 10.1 points below the typical model. That beats 1% of the ranked field. Among the 70 ranked Anthropic models it is #70. The next one up is claude-2.1 at 40.6.

claude-3-haiku-20240307, benchmark by benchmark

claude-3-haiku-20240307 lands below the middle on Aider Code Editing: 47.4%, #64 of 92, with 68% of the field scoring higher. Claude 3.5-Sonnet-20241022 leads at 84.2%; the closest result above is codestral:22b-v0.1-q8_0 at 48.1%. It covers Aider official benchmark. The board median is 55.6%, so claude-3-haiku-20240307 is 8.2 points below it. Source: Aider Aider Code Editing leaderboard, captured 2026-10-13.

GPQA Diamond is a weak spot. claude-3-haiku-20240307 scores 36.3%, #232 of 275, behind 84% of the models measured. The leader, GPT-6 Astra (max), reaches 95.8%. It covers expert-level science questions. The board median is 71.7%, so claude-3-haiku-20240307 is 35.4 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

DTBench is a weak spot. claude-3-haiku-20240307 scores 50.1%, #130 of 146, behind 88% of the models measured. The leader, Claude Opus 5.5 (max), reaches 98.9%. It covers dynamic tool-use benchmark. The board median is 68.8%, so claude-3-haiku-20240307 is 18.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

LMCA is a weak spot. claude-3-haiku-20240307 scores 8.8%, #98 of 108, behind 90% of the models measured. The leader, Claude Opus 5.5 (max), reaches 68.2%. It covers language-model capability assessment. The board median is 29.2%, so claude-3-haiku-20240307 is 20.4 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

OTIS Mock AIME is a weak spot. claude-3-haiku-20240307 scores 1.8%, #176 of 189, behind 93% of the models measured. The leader, Claude Fable 5.1 (max), reaches 100.0%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so claude-3-haiku-20240307 is 56.0 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

WeirdML is a weak spot. claude-3-haiku-20240307 scores 9.8%, #104 of 109, behind 94% of the models measured. The leader, GPT-6 Astra (max), reaches 93.3%. It covers unusual machine-learning tasks. The board median is 41.2%, so claude-3-haiku-20240307 is 31.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Strengths and weaknesses

Reasoning, maths and science: weak, ahead of 11% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#232 of 275) and its weakest is OTIS Mock AIME (#176 of 189). General knowledge and chat: weak, ahead of 7% of the field on average across 2 benchmarks. Its best board here is LMCA (#98 of 108) and its weakest is WeirdML (#104 of 109).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so claude-3-haiku-20240307 costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on GPT-3.5 Turbo (Jan 2024) ($0.500) and Claude 3.5 Haiku (Oct 2024) ($0.800).

Release history

The Anthropic model released before it was claude-3-sonnet-20240229 on 2024-03-04, which scores 41.9. The next one was Claude 3.5 Sonnet on 2024-06-20, at 49.2.

Reading these numbers

The 6 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (5), Aider Aider Code Editing leaderboard (1). 1 of them come from an official leaderboard and 5 from datasets or third-party runs. With 6 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#733 Llama-4-Scout-17B-16E-Instruct (40.1)
#734 gemma-3-4b-it (39.9)
#735 qwen2.5-7b-instruct (39.9)
#737 gemma-3-27b-it (39.7)
#738 Meta-Llama-3-8B-Instruct (39.4)
#739 gemma-3-12b-it (39.1)