Anthropic

Claude Haiku 4.5

Type
language model
Context
200K tokens
Max output
64K
Released
15 October 2025
Knowledge cutoff
February 2025
API string
claude-haiku-4-5-20251001

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
qihang-ai $0.14 $0.71 31 Jul 2026
Jiekou.AI $0.9 $4.5 31 Jul 2026
daoxe $1 $5 31 Jul 2026
freemodel $1 $5 31 Jul 2026
302.AI $1 $5 31 Jul 2026
Abacus.AI $1 $5 31 Jul 2026
Amazon Bedrock anthropic.claude-haiku-4-5-20251001-v1:0 $1 $5 2 Aug 2026
Anthropic $1 $5 31 Jul 2026
Helicone $1 $5 31 Jul 2026
LLM Gateway $1 $5 31 Jul 2026
Merge Gateway $1 $5 31 Jul 2026
NanoGPT $1 $5 31 Jul 2026
Amazon Bedrock apac.anthropic.claude-haiku-4-5-20251001-v1:0 $1.1 $5.5 2 Aug 2026
unorouter $1.2 $6 31 Jul 2026
Amazon Bedrock bedrock/us-gov-east-1/anthropic.claude-haiku-4-5-20251001-v1:0 $1.2 $6 2 Aug 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

A single provider appears several times if it sells this model under different identifiers and at different prices — for example, at different weight precision. The identifier is shown next to the name. Identical offers that arrived from two sources under different spellings are merged into one row.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score, programming 1,450.74 1,446–1,455
Arena Score, expert questions 1,443.71 1,437–1,451
Arena Score, long queries 1,426.81 1,423–1,431
Arena Score in Spanish 1,419.11 1,406–1,432
Arena Score, hard prompts 1,417.91 1,415–1,421
Arena Score in English 1,413.97 1,410–1,418
Arena Score, instruction following 1,413.07 1,409–1,417
Arena Score, multi-turn dialogue 1,407.02 1,402–1,412
Arena Score in French 1,406.71 1,393–1,420
Arena Score, overall 1,393.72 1,391–1,396
Arena Score, mathematics 1,392.77 1,385–1,401
Arena Score in Russian 1,376.88 1,370–1,383
Arena Score, creative writing 1,371.16 1,366–1,377
Arena Score, web development 1,325.04 1,320–1,330
MATH, difficulty level five 96.36 % thinking budget: 32K · with a tuned harness Epoch evaluations
Mock AIME 2024–2025 — olympiad problems 66.63 % thinking budget: 32K · with a tuned harness Epoch evaluations
GPQA Diamond — graduate-level questions 61.62 % thinking budget: 32K · with a tuned harness Epoch evaluations
ARC-AGI — generalising to unseen patterns 47.67 % thinking budget: 32K · with a tuned harness
WeirdML — unusual machine learning tasks 45.4 % WeirdML Leaderboard
Terminal-Bench — working in the command line 35.5 % · with a tuned harness Terminal-Bench v2 Leaderboard
BALROG — game environments 31.2 % Balrog Leaderboard
APEX-Agents 8.9 %
SimpleQA Verified — factual accuracy 5.9 % thinking budget: 32K Epoch evaluations
ARC-AGI-2 4.03 % thinking budget: 32K
Chess puzzles 3.2 % thinking budget: 32K Epoch evaluations
CritPt — physics problems 0 %