[mle]/DaBench
Leaderboard

Claude Haiku 5.5

anthropic/claude-haiku-5.5

Hostedanthropic225 tok/sopenrouterserved by Amazon Bedrock1M context
Run recordCompare
Overall score
70

113 of 153 test cases fully passed

Scores by task

Scores give partial credit. Passed counts fully correct answers only. Select a task to read its test cases.

  1. 61Agents12 of 21 passedp90 latency 1m 5s
  2. 90Structured Extraction19 of 21 passedp90 latency 23.1s
  3. 86Coding18 of 21 passedp90 latency 54.8s
  4. 76Grounded Answers16 of 21 passedp90 latency 23.4s
  5. 94Instruction Following19 of 21 passedp90 latency 47.8s
  6. 67Business Logic14 of 21 passedp90 latency 54.8s
  7. 67Security & permissions14 of 21 passedp90 latency 37.1s
  8. 17Diagnosis & investigation1 of 6 passedp90 latency 40.6s

What it built

What it wrote for showcase prompts, running live in your browser.

See everything it built

Run it yourself

The API call the benchmark made, and the settings behind this score.

Deployment
Hosted
API
OpenRouter
Serving provider
Amazon Bedrock
Precision
Provider default
Context window
1M
Throughput
225 tok/s
Cost of a suite run
$0.42
Cost per case
$0.0028
Time per case
25.5s
Measured
2026-10-07
Run this buildOpenRouterHosted
curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"anthropic/claude-haiku-5.5","provider":{"order":["amazon-bedrock"],"allow_fallbacks":false},"service_tier":"default","messages":[{"role":"user","content":"Hello"}]}'

The request the benchmark sent, pinned to the provider that served it.

One email when a new model lands on the board or in the showcase. Unsubscribe anytime.

LeaderboardCompare Claude Haiku 5.5 with another model