Skip to content
aiqbench

Anthropic · model passport

Anthropic: Claude Haiku 4.5

claude-haiku-4-5
  • Cloud only
  • Proprietary licence
  • 200,000 context
  • Released 2025-10-15(curated)
16/27axes covered
Overall · LMArena
1,414#131 of 396
Best percentile
93.9BFCL v4 (function calling)
Cohort standing
48.4median 53.2

What the data says

Benchmarked on 16 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 93.9 pct — BFCL v4 (function calling) (68.7%)

Weakest axes

  • 7.9 pct — Document Arena (1,419)
  • 9.8 pct — SimpleQA Verified (13.2%)

Uneven: 86 percentile points between its best and worst axis — 93.9 pct on BFCL v4 (function calling) versus 7.9 pct on Document Arena.

  • BFCL v4 (function calling)Dormant board93.9 pct68.7%Verify →
  • Coding75.4 pct1,482Verify →
  • Instruction following73 pct1,415Verify →
  • LiveBench72.9 pct65.0%Verify →
  • Hard prompts69.9 pct1,441Verify →
  • Overall67.2 pct1,414Verify →
  • GPQA Diamond52.9 pct71.2%Verify →
  • AIME (math)47.5 pct66.7%Verify →
  • HHEM Hallucination (Vectara)45.7 pct90.2%Verify →
  • Terminal-Bench38.6 pct35.5%Verify →
  • FrontierMath37.7 pct5.9%Verify →
  • ARC-AGI-229.2 pct4.0%Verify →
  • Factuality Arena28.5 pct1,418Verify →
  • WebDev Arena19.4 pct1,330Verify →
  • SimpleQA Verified9.8 pct13.2%Verify →
  • Document Arena7.9 pct1,419Verify →
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • SWE-bench VerifiedNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Use it hosted

API model IDanthropic/claude-haiku-4.5
Context length200,000 tokens
Max output64,000 tokens
Acceptstext, image, file
Input price$1 per 1M tokens
Cached input price$0.10 per 1M tokens
Output price$5 per 1M tokens

API parameters this endpoint supports

Include reasoningMax completion tokensMax tokensReasoningResponse formatStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"anthropic/claude-haiku-4.5","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • claude-haiku-4-5-20251001 stands at 48.4 (raw mean 48.1) — measured on 16 of 27 axes.

    standingtemplateVerify →

  • claude-haiku-4-5-20251001 stands at 48.4 versus a cohort median of 53.2 across 133 models released within 180 days — below the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →