claude-haiku-4-5-20251001: benchmark scores, price and ranking
claude-haiku-4-5-20251001 from Anthropic ranks #602 of 743 LLMs on the LLMs Tiger aggregate benchmark score (45.2), based on 8 benchmark results captured up to 2026-10-13.
- Organisation
- Anthropic
- Type
- Fast / low-cost (flash) model
- Open weights
- No
- Released
- 2025-10-15 (source: curated)
- Context window
- 200,000 tokens
- Cheapest API price
- $1.00 in / $5.00 out per million tokens (via anthropic, observed 2026-10-13T19:47Z)
- Aggregate rank
- #602 of 743 (score 45.2)
Relative to other models, claude-haiku-4-5-20251001 is strongest on WeirdML (#41 of 109) and weakest on SimpleQA Verified (#62 of 69).
Across the 8 benchmarks where at least ten models are scored, claude-haiku-4-5-20251001 is in the top 10% on 0 and in the top half on 2.
Its cheapest listed input price, $1.00 per million tokens, is lower than that of 34% of the 427 models with a listed price.
claude-haiku-4-5-20251001 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| WeirdML | 45.4% | #41 of 109 | Epoch AI Benchmarking Hub | 2026-10-13 |
| BALROG | 31.2% | #13 of 34 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 60.5% | #166 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 35.8% | #117 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Terminal-Bench 2.0 | 29.8% | #19 of 28 | Epoch AI Benchmarking Hub | 2026-10-13 |
| LMArena Document Elo | 1,420 | #14 of 18 | LMArena | 2026-10-13 |
| WebDev Arena | 1,286 | #55 of 67 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleQA Verified | 13.2% | #62 of 69 | Epoch AI Benchmarking Hub | 2026-10-13 |
About claude-haiku-4-5-20251001
claude-haiku-4-5-20251001 is a smaller, lower-cost model from Anthropic. It was released on 2025-10-15, according to the developer's announcement. It accepts up to 200,000 tokens of input at once. The cheapest API listing we track charges $1.00 per million input tokens and $5.00 per million output tokens, through anthropic. We hold 8 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where claude-haiku-4-5-20251001 stands
On our aggregate score, claude-haiku-4-5-20251001 is #602 of 743 ranked models, in the lower half. Its score is 45.2; the scale is centred on 50, so it sits 4.8 points below the typical model. That beats 19% of the ranked field. Among the 70 ranked Anthropic models it is #62. The next one up is Claude 3.5 Sonnet (Oct 2024) at 45.4. The next one down is Claude 3.5 Sonnet (Jun 2024) at 44.2.
claude-haiku-4-5-20251001, benchmark by benchmark
WeirdML puts claude-haiku-4-5-20251001 at 45.4%, #41 of 109. That beats 62% of the models tested, but 47.9 points short of GPT-6 Astra (max) (93.3%). The results either side are Kimi K2.5 (45.6%) and claude-haiku-4-5-20251001_16K (44.1%). It covers unusual machine-learning tasks. The board median is 41.2%, so claude-haiku-4-5-20251001 is 4.2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On BALROG, claude-haiku-4-5-20251001 reaches 31.2%. It ranks #13 of 34, ahead of 62% of the field, and 37.1 points behind GPT-6 Astra (max) at 68.3%. The results either side are GPT-4o (May 2024) (32.3%) and claude-haiku-4-5-20251001_1K (31.2%). It covers long-horizon game-playing agents. The board median is 27.6%, so claude-haiku-4-5-20251001 is 3.6 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-haiku-4-5-20251001 lands below the middle on GPQA Diamond: 60.5%, #166 of 275, with 60% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is Qwen3-30B-A3B at 61.7%. It covers expert-level science questions. The board median is 71.7%, so claude-haiku-4-5-20251001 is 11.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-haiku-4-5-20251001 lands below the middle on OTIS Mock AIME: 35.8%, #117 of 189, with 61% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is GPT-4.5 Preview (Feb 2025) at 37.8%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so claude-haiku-4-5-20251001 is 21.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-haiku-4-5-20251001 lands below the middle on Terminal-Bench 2.0: 29.8%, #19 of 28, with 64% of the field scoring higher. Gemini 3 Pro Preview leads at 69.4%; the closest result above is MiniMax-M2 at 30.0%. It covers real terminal tasks (agent + model). The board median is 37.3%, so claude-haiku-4-5-20251001 is 7.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
claude-haiku-4-5-20251001 lands below the middle on LMArena Document Elo: 1,420, #14 of 18, with 72% of the field scoring higher. Claude Fable 5.1 (max) leads at 1,513; the closest result above is gemma-4-31b at 1,425. It covers crowd-voted questions about documents. The board median is 1,440, so claude-haiku-4-5-20251001 is 20 points below it. Source: LMArena, captured 2026-10-13.
WebDev Arena is a weak spot. claude-haiku-4-5-20251001 scores 1,286, #55 of 67, behind 81% of the models measured. The leader, Claude Opus 5.5 (max), reaches 1,820. It covers building web apps, crowd-voted (Elo). The board median is 1,433, so claude-haiku-4-5-20251001 is 147 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
SimpleQA Verified is a weak spot. claude-haiku-4-5-20251001 scores 13.2%, #62 of 69, behind 88% of the models measured. The leader, gemini-4-argon-high, reaches 76.0%. It covers short factual questions. The board median is 41.0%, so claude-haiku-4-5-20251001 is 27.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Software engineering: weak, ahead of 25% of the field on average across 2 benchmarks. Its best board here is Terminal-Bench 2.0 (#19 of 28) and its weakest is WebDev Arena (#55 of 67). Reasoning, maths and science: uneven, ahead of 39% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#166 of 275) and its weakest is OTIS Mock AIME (#117 of 189). General knowledge and chat: uneven, ahead of 32% of the field on average across 3 benchmarks. Its best board here is WeirdML (#41 of 109) and its weakest is SimpleQA Verified (#62 of 69).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so claude-haiku-4-5-20251001 costs more 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 mimo-v2-pro ($1.10) and o3-mini-2025-01-31 (low) ($1.10).
Release history
The Anthropic model released before it was Claude 4.5 Haiku on 2025-10-15, which scores 50.3. The next one was claude-haiku-4-5-20251001_32K on 2025-10-15, at 47.3.
Reading these numbers
The 8 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (7), LMArena (1). No official leaderboard supplied any of them, so treat them as third-party measurements.
Models ranked nearby
- #599 Nemotron 3 Super (45.3)
- #600 Llama-2-70b-chat (45.3)
- #601 Mistral-Nemo-Instruct-2407 (45.2)
- #603 Meta-Llama-3-8B (45.2)
- #604 deepseek-math-7b-instruct (45.2)
- #605 ollama/hermes3:8b-llama3.1-fp16 (45.1)
More from Anthropic
- Claude Opus 5.5 (max) (#1, 64.3)
- Claude Fable 5.1 (max) (#3, 61.9)
- Claude Sonnet 5.5 (max) (#4, 60.8)
- Claude Opus 5 (max) (#8, 60.0)
- Claude Fable 5 (max) (#9, 59.9)
- claude-opus-4-8-xhigh-effort (#14, 56.6)
Compare it on the live leaderboard → · How the aggregate score works