MDL
Claude 3.5 Haiku
Anthropic
Verified against Artificial Analysis · Jul 13, 2026
Capability
46
percentile index
Price /M
$0.8 / $4
in / out
Context
200K
tokens
Avg score
38.4
12 benchmarks
AA Intelligence
12.3
index
Weights
Closed
Released
Oct 22, 2024
About
Claude 3.5 Haiku is a proprietary model in the Claude 3.5 family from Anthropic. Benchmarked on 12 evals, averaging 38.4.
proprietarytext
Benchmark scores · 12
avg 38.4 — higher is betterThe research behind this model
all papers mentioning it →Qwen2.5-VL Technical Report52 cites · Feb 19, 2025Tulu 3: Pushing Frontiers in Open Language Model Post-Training7 cites · Nov 22, 2024Evolutionary Dynamics of Cooperation in Next-Generation LLM Agent Systems: A Cross-Provider Empirical ExtensionMay 28, 2026Findings of the Counter Turing Test: AI-Generated Text DetectionMay 20, 2026Multilingual jailbreaking of LLMs using low-resource languagesMay 18, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Details
PublisherAnthropic
FamilyClaude 3.5
Parameters—
Licenseproprietary
WeightsClosed
ReleasedOct 22, 2024
Performance & price
Price in / out$0.8 / $4 per 1M
Blended price$1.6 / 1M
Coding index15.9
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Anthropic200K ctx · $4/M
Artificial Analysis$4/M