Compare

Claude Haiku 4.5 vs GPT-5.4 Mini

Pricing per million tokens, context window, capabilities — pulled from each provider's public docs. All 2 are available via the same AIgateway OpenAI-compatible endpoint; flip the model string to switch.

Search2/4
Claude Haiku 4.5
anthropic/claude-haiku-4.5
GPT-5.4 Mini
openai/gpt-5.4-mini
Provider
Anthropic
OpenAI
Family
Claude 4
GPT-5
Modality
text
text
Context window
200,000 tok
128,000 tok
Max output
8,192 tok
16,384 tok
Released
2026-05-22
2026-05-22
License
Proprietary
Proprietary
Input price
$1.00 /1M
$0.750 /1M
Output price
$5.00 /1M
$4.50 /1M
Cache read
$0.100 /1M
$0.075 /1M
Cache write
$1.25 /1M
Tools
yes
yes
Streaming
yes
yes
Vision
yes
yes
JSON mode
yes
yes
Reasoning
yes
Prompt caching
yes
yes
Batch API
yes
Try it
Open in playground →
Open in playground →
Claude Haiku 4.5
anthropic/claude-haiku-4.5
Full spec →

Claude Haiku 4.5 delivers similar levels of coding performance at one-third the cost and more than twice the speed of larger models.

Strengths
  • Fastest Claude
  • Cheap enough for per-token workloads
  • Still supports tools + vision
Use cases
ClassifiersRoutingHigh-throughput chatReal-time agents
GPT-5.4 Mini
openai/gpt-5.4-mini
Full spec →

GPT-5.4 mini is a smaller, faster, and more cost-efficient version of GPT-5.4 for lightweight tasks.

Strengths
  • Very cheap per token
  • Fast
  • Full tool + vision support
Use cases
Drafts + rewritesBulk batch jobsCost-sensitive production

Benchmarks

Claude Haiku 4.5
GPT-5.4 Mini
HumanEval
85.2
88.6
MMLU
80.1
84.5

Source: each provider's published benchmarks. Higher is better. Run an eval to compare on your own data.

Compare with another

Claude Sonnet 4.6 vs GPT-5.4 Mini
anthropic/claude-sonnet-4.6 · openai/gpt-5.4-mini
Claude Haiku 4.5 vs Gemini 3 Flash
anthropic/claude-haiku-4.5 · google/gemini-3-flash
GPT-5.4 Mini vs Gemini 3 Flash
openai/gpt-5.4-mini · google/gemini-3-flash
Claude Opus 4.8 vs GPT-5.4 Mini
anthropic/claude-opus-4.8 · openai/gpt-5.4-mini
Claude Opus 4.8 vs Claude Haiku 4.5
anthropic/claude-opus-4.8 · anthropic/claude-haiku-4.5
MiniMax M3 vs GPT-5.4 Mini
minimax/m3 · openai/gpt-5.4-mini
MiniMax M3 vs Claude Haiku 4.5
minimax/m3 · anthropic/claude-haiku-4.5
Gemini 3.1 Pro vs GPT-5.4 Mini
google/gemini-3.1-pro · openai/gpt-5.4-mini
Gemini 3.1 Pro vs Claude Haiku 4.5
google/gemini-3.1-pro · anthropic/claude-haiku-4.5
SWITCH BETWEEN THEM

One key, all 2, one line different.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)

# Claude Haiku 4.5
client.chat.completions.create(
    model="anthropic/claude-haiku-4.5",
    messages=[{"role":"user","content":"hello"}],
)

# GPT-5.4 Mini
client.chat.completions.create(
    model="openai/gpt-5.4-mini",
    messages=[{"role":"user","content":"hello"}],
)
Get an AIgateway keyRun an eval on these →