Anthropic
Claude Haiku 4.5
model
claude-haiku-4-5MidStableVisionStreamingVisionToolsBatchLong contextUnited States
Context
200K
Input / 1M tokens
$1.00
Output / 1M tokens
$5.00
Weighted tokens formula
1× provider × 4× tier
About this model
Claude Haiku 4.5
Claude Haiku 4.5 — Anthropic's fastest + cheapest. Speed-optimized for high-throughput chat, classification, and real-time UX while preserving Anthropic's safety + alignment tuning.
Released
2024-12
Training cutoff
2024-11
Parameters
Not disclosed
Best for
Use cases
- ✓High-volume customer chat
- ✓Classification + routing
- ✓Real-time voice agent backbone
- ✓Cheap-tier model for cost-sensitive tiers
Strengths
What it does well
- ●Sub-second TTFT typical
- ●Strong instruction-following for the size class
- ●200K context window
- ●Same safety tuning as Sonnet/Opus
Limitations
Know the trade-offs
- !Reasoning depth limited vs Sonnet
- !Code generation trails Qwen 2.5 Coder
- !Long-document comprehension weaker on tasks past 50K tokens
Code example
Chat completion with Claude Haiku 4.5
from openai import OpenAI
client = OpenAI(
base_url="https://api.anoman.io/v1",
api_key="anm-sk-..."
)
response = client.chat.completions.create(
model="claude-haiku-4-5",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)Weighted tokens
weighted_tokens = raw_tokens × 1 (provider) × 4 (tier)
Pro plan: 20M weighted tokens/month. Combined multiplier 4×: 5,000,000 raw tokens available.
Alternatives
Similar models in our catalog
Workloads that fit Claude Haiku 4.5 often fit these too. Compare benchmarks and pricing before locking in a default.