GetInfoUs
Head-to-Head Model Comparison

Claude 3.5 Haiku vs Claude 3 Haiku

Benchmarks, API Token Pricing, Context Window & Speed Comparison

Geprüft vom GetInfoUs-ForschungsteamZuletzt geprüft: October 2026Redaktionelle Richtlinien
Teilen
Related topics (12)
AI CompareAI BenchmarkToken PricingModel CostModel SpecsSWE-bench CodingModel VersusWhich AI ModelLLM Speed & LatencyReasoning ModelsExtended ThinkingPrompt Caching ROI
Claude 3.5 Haiku vs Claude 3 Haiku: Benchmarks, Pricing & Speed Comparison

Claude 3.5 Haiku von Anthropic und Claude 3 Haiku von Anthropic gehören zu den führenden KI-Modellen. Claude 3.5 Haiku bietet ein Kontextfenster von 200K Tokens bei API-Preisen von $0.80/$4.00 pro 1M Tokens, während Claude 3 Haiku 200K Tokens Kontext zu $0.25/$1.25 pro 1M Tokens bereitstellt. Hier finden Sie den vollständigen Benchmark-, Latenz- und Kostenvergleich.

Direktvergleich (VS)

Beliebige Modelle wechseln & vergleichen

Select any AI models from our verified benchmark database to generate live side-by-side matrices.

Profile
Profile
Multi-Model Arena Stack:
5-Model Matrix Workbench
Popular Matchups:
Anthropic

Claude 3.5 Haiku

Autonomes Agenten-Modell

Claude 3.5 Haiku von Anthropic ist ein KI-Modell für autonome Agenten-Workflows, zuverlässiges Tool-Calling und mehrstufige Automatisierung mit 200K Kontextfenster, 130 Token/s Durchsatz, einem Intelligenz-Index von 36/100 und $0.8/$4 pro 1

Intelligence Index36 / 100
Input Token:

$0.80 / 1M

Output Token:

$4.00 / 1M

Context Window:

200K tokens

SWE-bench Verified:

40.6%

Anthropic

Claude 3 Haiku

Ultraschnelle Echtzeit-Inferenz

Claude 3 Haiku von Anthropic ist ein KI-Modell für ultraschnelle Echtzeit-Antworten, niedrige Latenz und Hochdurchsatz-Routing mit 200K Kontextfenster, 125 Token/s Durchsatz, einem Intelligenz-Index von 33/100 und $0.25/$1.25 pro 1M Token.

Intelligence Index33 / 100
Input Token:

$0.25 / 1M

Output Token:

$1.25 / 1M

Context Window:

200K tokens

SWE-bench Verified:

10.5%

Wichtigste Unterschiede auf einen Blick

  • Claude 3 Haiku is 3.2x cheaper for API usage (input + output tokens combined).
  • Claude 3.5 Haiku is faster at 130 tok/s vs 125 tok/s.
  • Claude 3.5 Haiku leads on SWE-bench coding (40.6% vs 10.5%).
Need definitions for token pricing, context, or benchmarks?Explore AI Glossary →
FeatureClaude 3.5 HaikuClaude 3 Haiku
Intelligence Index
Composite reasoning & knowledge score (0-100)
3633
Provider
AnthropicAnthropic
Input Price
Cost per million input tokens
$0.80 / 1M$0.25 / 1M
Output Price
Cost per million output tokens
$4.00 / 1M$1.25 / 1M
Blended Price
Weighted 7:2:1 cache:input:output per 1M
$0.62 / 1M$0.19 / 1M
Context Window
Maximum input + output tokens
200K tokens~300 A4 pages of text200K tokens~300 A4 pages of text
Max Output Tokens
Per single request
8K4K
Throughput Speed
Tokens generated per second
130 tok/s125 tok/s
SWE-bench Verified
Real-world software engineering
40.6%10.5%
GPQA Diamond
Graduate-level reasoning
41.6%33%
Cached Input Price
Prompt caching discount per 1M tokens
$0.08 / 1M$0.03 / 1M

Wer gewinnt in welcher Disziplin?

At-a-glance decision framework
Coding
Claude 3.5 Haiku
+30.1% SWE-bench lead
Cost
Claude 3 Haiku
3.2x cheaper monthly runtime
Speed
Claude 3.5 Haiku
+4% faster generation
Context
Tied
200K tokens window

Geprüfte Benchmark-Ergebnisse im Vergleich

Official evaluation metrics verified by independent third-party research labs.

Claude 3.5 HaikuClaude 3 Haiku
SWE-bench Verified (Autonomous Coding)

Industry benchmark evaluating real-world GitHub issue resolution and multi-file debugging.

40.6%vs10.5%
GPQA Diamond (Doctorate-Level Science & Math)

Challenging questions created by domain experts to test complex PhD-level reasoning.

41.6%vs33%
MMLU-Pro (Multi-task Language Understanding)

Hardened reasoning and academic knowledge benchmark with reduced lucky-guess probability.

69.2%vs61.2%
LMSYS Chatbot Arena Elo

Crowdsourced human preference evaluation based on blind, head-to-head model battles.

1260ptsvs1175pts
Humanity's Last Exam (Frontier Expert Knowledge)

Extremely difficult, closed-ended questions spanning the frontier of human expert knowledge.

6.7%vsN/A
Terminal-Bench (Agentic Coding & Terminal Use)

Autonomous multi-step tasks executed inside a real command-line terminal environment.

26.8%vsN/A
SciCode (Scientific Computing)

Research-grade coding problems drawn from real scientific and numerical workloads.

24.4%vsN/A
LiveCodeBench (Competitive Coding)

Contamination-free competitive programming problems collected continuously over time.

43.6%vsN/A
AIME (Advanced Math Competition)

American Invitational Mathematics Examination — multi-step Olympiad-level reasoning.

47.8%vsN/A
Long-Context Reasoning

Accuracy when reasoning over very large inputs that fill the model context window.

62.3%vsN/A
Generation Speed (Throughput)

Tokens emitted per second during active generation streaming.

130 tok/s vs125 tok/s

Latenz & Antwortzeit

Time to first token (TTFT) and estimated end-to-end response time for a typical 500-token response.

Time to First Token (TTFT)Lower is better
Claude 3.5 Haiku0.38s
Claude 3 Haiku0.35s
Est. End-to-End Response (500 tokens)TTFT + generation time
Claude 3.5 Haiku4.23s
Claude 3 Haiku4.35s

Interaktiver Token-Kosten-Simulator

Adjust sliders to calculate projected monthly cloud API spending with prompt caching discounts.

100K / day
10K500K1M
30K / day
5K150K300K
50% cached
0% (Cold)50%90% (Warm)
Claude 3.5 Haiku Monthly API Bill
$4.92 / month
$0.8/1M in · $4/1M out
Claude 3 Haiku Monthly API Bill
$1.54 / month
$0.25/1M in · $1.25/1M out
Claude 3 Haiku saves approximately $3.38 / month (69%) under this workload.3.2x Savings

Claude 3.5 Haiku · Strengths & Key Advantages

  • 200K Kontextfenster und bis zu 8K maximale Output-Token (SWE-bench Verified 40.6% · GPQA Diamond 41.6%)
  • Hohe Generierungsgeschwindigkeit von 130 Token/Sekunde bei nur 380 ms Erstlatenz (TTFT)
  • Effiziente API-Struktur ($0.8/$4 pro 1M Token) inklusive Prompt-Caching, Tool-Calling und strikter JSON-Ausgabe
Considerations & Trade-offs
  • Proprietäres Cloud-Modell ohne Möglichkeit zum lokalen Download der Modellgewichte
  • Bei extrem komplexen mehrstufigen Forschungsfragen bieten schwere Reasoning-Spezialmodelle noch tiefere Analysen

Claude 3 Haiku · Strengths & Key Advantages

  • 200K Kontextfenster und bis zu 4K maximale Output-Token (SWE-bench Verified 10.5% · GPQA Diamond 33%)
  • Hohe Generierungsgeschwindigkeit von 125 Token/Sekunde bei nur 350 ms Erstlatenz (TTFT)
  • Effiziente API-Struktur ($0.25/$1.25 pro 1M Token) inklusive Prompt-Caching, Tool-Calling und strikter JSON-Ausgabe
Considerations & Trade-offs
  • Proprietäres Cloud-Modell ohne Möglichkeit zum lokalen Download der Modellgewichte
  • Bei extrem komplexen mehrstufigen Forschungsfragen bieten schwere Reasoning-Spezialmodelle noch tiefere Analysen

Token Use & Task Economics

Estimated cost and token consumption to complete one representative agentic task.

MetricClaude 3.5 HaikuClaude 3 Haiku
Cost per Task
Estimated USD to finish one task
$0.90$0.20
Output Tokens / Task
Tokens generated per task
45K24K
Reasoning Tokens / Task
Thinking tokens burned per task
0K0K

Technische Spezifikationen

Architecture, release timeline, capabilities, and modalities.

FeatureClaude 3.5 HaikuClaude 3 Haiku
Official API Model ID (SDK)claude-3-5-haiku-20241101claude-3-haiku-20240301
API ProtocolAnthropic Messages API + AWS Bedrock + Vertex AIAnthropic Messages API + AWS Bedrock + Vertex AI
Release Date2024-112024-03
Knowledge CutoffJuly 2024August 2023
ArchitectureOptimized Lightweight TransformerCompact High-Throughput Transformer
ParametersConfidentialConfidential
Enterprise Rate Limits (RPM / TPM)4,000 – 10,000 RPM (Tier 4 + Priority Tier) · 800,000 – 2,000,000 TPM4,000 – 10,000 RPM (Tier 4 + Priority Tier) · 800,000 – 2,000,000 TPM
Long-Context Tier (>128K/200K)Standard rate up to 200K tokens; 2x input/output tier above 200K tokensStandard rate up to 200K tokens; 2x input/output tier above 200K tokens
Prompt Cache Policy & TTL$0.08/1M cached read (90% off); 5-min default & 1-hour extended TTL$0.025/1M cached read (90% off); 5-min default & 1-hour extended TTL
Computer Use / GUI AgentSupported (Screen coordinate grounding, browser & desktop GUI action loops)Supported (Screen coordinate grounding, browser & desktop GUI action loops)
MCP & Parallel Tool CallingNative MCP Client/Server ready · Parallel function calling · 100% Strict JSON SchemaNative MCP Client/Server ready · Parallel function calling · 100% Strict JSON Schema
Web Search Grounding & SandboxNative Web Search Citations + Code Execution Tool + MCP ConnectorsNative Web Search Citations + Code Execution Tool + MCP Connectors
Official Fine-Tuning SupportCustom Model Tuning via Amazon Bedrock & Prompt DistillationCustom Model Tuning via Amazon Bedrock & Prompt Distillation
Multimodal Token BillingText BPE (~4 chars/token) · Vision (~258–765 tokens per image tile)Text BPE (~4 chars/token) · Vision (~258–765 tokens per image tile)
Quantization & Min GPU VRAMCloud-Hosted Proprietary (FP8 / BF16 Tensor Core Cluster) · Zero Local VRAM Required (100% Serverless Cloud API)Cloud-Hosted Proprietary (FP8 / BF16 Tensor Core Cluster) · Zero Local VRAM Required (100% Serverless Cloud API)
Safety & Frontier ClassificationAnthropic RSP ASL-2 / ASL-3 Constitutional Safety EvaluatedAnthropic RSP ASL-2 / ASL-3 Constitutional Safety Evaluated
Enterprise Compliance (ZDR/HIPAA/SOC2)SOC2 Type II · GDPR · Zero Data Retention (ZDR) & HIPAA BAA Available on Enterprise TierSOC2 Type II · GDPR · Zero Data Retention (ZDR) & HIPAA BAA Available on Enterprise Tier
Input Modalitiestext, visiontext, vision
Output Modalitiestexttext

Claude 3.5 Haiku vs Claude 3 Haiku: Schnelle Antworten (FAQ)

Which is more intelligent, Claude 3.5 Haiku or Claude 3 Haiku?

Claude 3.5 Haiku is more intelligent. It scores 36 compared with 33 on the composite Intelligence Index.

Which model is faster?

Claude 3.5 Haiku is faster, generating 130 tokens per second compared with 125 tokens per second.

Which model is cheaper to run?

Claude 3 Haiku is cheaper at $0.19 per 1M tokens blended, compared with $0.62 (7:2:1 cache:input:output ratio).

Which model has lower latency?

Claude 3 Haiku has lower latency, with a time to first token of 0.35s compared with 0.38s.

Fazit

Sowohl Claude 3.5 Haiku (Optimal für autonome Agenten-Workflows, zuverlässiges Tool-Calling und mehrstufige Automatisierung, Enterprise-RAG-Systeme und skalierbare API-Pipelines.) als auch Claude 3 Haiku (Optimal für ultraschnelle Echtzeit-Antworten, niedrige Latenz und Hochdurchsatz-Routing, Enterprise-RAG-Systeme und skalierbare API-Pipelines.) überzeugen im Produktionseinsatz: Claude 3.5 Haiku führt bei komplexem Reasoning und Coding, Claude 3.5 Haiku liefert den höheren Token-Durchsatz und Claude 3 Haiku ist die kosteneffizienteste Wahl für große API-Workloads.

Häufig gestellte Fragen (FAQ)

Welches Modell ist besser für Coding und Reasoning – Claude 3.5 Haiku oder Claude 3 Haiku?+

Im Gesamt-Intelligenzindex und bei Software-Benchmarks liegt Claude 3.5 Haiku vorn. Prüfen Sie oben die SWE-bench- und GPQA-Werte für Ihren Anwendungsfall.

Wie unterscheiden sich die API-Preise von Claude 3.5 Haiku und Claude 3 Haiku?+

Claude 3.5 Haiku kostet $0.80 / $4.00 pro 1M Input-/Output-Tokens, Claude 3 Haiku liegt bei $0.25 / $1.25. Beim gewichteten Gesamtpreis ist Claude 3 Haiku günstiger.

Welches Modell hat das größere Kontextfenster und die höhere Geschwindigkeit?+

Claude 3.5 Haiku bietet 200K Tokens Kontext gegenüber 200K Tokens bei Claude 3 Haiku. Bei der Generierungsgeschwindigkeit liegt Claude 3.5 Haiku vorn.

AI Terminology & Guide
Explore Glossary →

Understand key metrics like context window, token pricing, reasoning mode, and SWE-bench scores in our comprehensive AI glossary.

Tokens & PricingContext WindowAI BenchmarksReasoning ModelsTool CallingModel Context Protocol (MCP)

Related reading

Benchmark scores, token prices, and rate limits evolve rapidly as AI labs release model updates. Always verify current specifications and pricing on the official provider documentation.

Offizielle Quellen: Offizielle API-Dokumentation von Anthropic und Anthropic, Systemkarten und unabhängige Benchmark-Auswertungen (2026).