VerdictPal
Join

Models / Anthropic

Claude Haiku 4.5

Claude Haiku · Released 2025-10-01

Anthropic's speed tier. 200K context, 64K output, $1/$5 per 1M tokens. Best for high-volume, low-latency tasks.

Fast, cheap Claude for routing, classification, extraction, and high-volume backend tasks.

#22 of 24 on Vals Index · current snapshot
40 Vals Index · 7 benchmark rows · 7 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite61
Context200K
Input / 1M$1.0
Output / 1M$5.0
Knowledge cutoff2025-08-01
Statussolid

Family profile

Best published score in each covered benchmark family.

66/100 avg
Knowledge
86
Reasoning
58
Agentic
61
Long context
45
Tool use
72
Human preference
74

6 tested benchmark families

Benchmark placements

Where Claude Haiku 4.5 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
LMArena (Chatbot Arena)Human preference#13/ 1374LMArena leaderboard2026-05-25
BFCLTool use#19/ 2072Berkeley Function Calling Leaderboard V42026-04-15
LongBench v2Long context#19/ 1945LongBench v2 leaderboard2026-04-01
LiveBenchReasoning#20/ 2158LiveBench leaderboard2026-04-15
MMLUKnowledge#20/ 2286Anthropic Claude Haiku 4.52025-10-01
Vals IndexAgentic#21/ 2340Vals AI — Vals Index2026-06-04
Terminal-BenchAgentic#65/ 18461Terminal-Bench 2.0 leaderboard2026-06-09
Family context

The three highest-scoring models with pages in each capability family. Where Claude Haiku 4.5 shows up, it's highlighted.

KnowledgeClaude Haiku 4.5 · 86
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningClaude Haiku 4.5 · 58
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
AgenticClaude Haiku 4.5 · 61
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextClaude Haiku 4.5 · 45
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Tool useClaude Haiku 4.5 · 72
  1. 1
    Claude Sonnet 4.6
    89
  2. 2
    GPT 5.5
    87
  3. 3
    Claude Opus 4.8
    87
Human preferenceClaude Haiku 4.5 · 74
  1. 1
    GPT 5.5
    95
  2. 2
    Claude Opus 4.8
    92
  3. 3
    Gemini 3.5 Flash
    91
Newest receipts

Benchmark rows added to the public ledger for Claude Haiku 4.5 in the last 120 days. Older rows live in the full table below.

3 recent
BenchmarkFamilyScoreSourceDays ago
Terminal-BenchAgentic61.4%Terminal-Bench 2.0 leaderboard89
Vals IndexAgentic40.33%Vals AI — Vals Index94
LMArena (Chatbot Arena)Human preference1368LMArena leaderboard104
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge86
Reasoning58
Agentic61
Long context45
Tool use72
Human preference74
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Claude Haiku 4.5. Scores link to the original source; gaps mean no public row exists yet.

7 sourced rows

7 of 37 catalog benchmarks have a sourced row for Claude Haiku 4.5.

Knowledge

Reasoning

Agentic

Long context

Tool use

Human preference

Changelog
  1. Released Claude Haiku 4.5 — speed tier, $1/$5, 200K context.
Sources