Benchmark Almanac

Credit Firewall Benchmark

Claude Haiku 4.5 100% leads since October 2026
Not in model cards

Models decide whether evidence still authorizes a credit decision after roles, targets, timing, or conversation context change.

Kaggle publishes scores for 12 distinct model versions.

Released
October 2026
Kind
Deception and alignment
Level
University
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
In views
Law and finance
Strongest model tested
Claude Opus 5.5 #5 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
12
Models scored on its own leaderboard

11 models share the top score of 100% among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude Haiku 4.5 100%
  2. 2 Claude Opus 4.5 100%
  3. 3 GPT-5.4 100%
  4. 4 GPT-5.4 Mini 100%
  5. 5 GPT-5.4 Nano 100%
  6. 6 GPT-5.5 100%
  7. 7 GPT-5.6 Sol 100%
  8. 8 Gemini 3.8 Flash 100%
  9. 9 Claude Opus 5.5 100%
  10. 10 GPT-6 Luna 100%
  11. 11 GPT-6 Sol 100%
  12. 12 Gemini 3.6 Flash 0%

Source: Official Credit Firewall Benchmark leaderboard

#ModelScoreSettingReleasedRun onSource
1 Claude Haiku 4.5 100% Oct 2025 Oct 2026 Official leaderboard
2 Claude Opus 4.5 100% Nov 2025 Oct 2026 Official leaderboard
3 GPT-5.4 100% Mar 2026 Oct 2026 Official leaderboard
4 GPT-5.4 Mini 100% Mar 2026 Oct 2026 Official leaderboard
5 GPT-5.4 Nano 100% Mar 2026 Oct 2026 Official leaderboard
6 GPT-5.5 100% Apr 2026 Oct 2026 Official leaderboard
7 GPT-5.6 Sol 100% Jul 2026 Oct 2026 Official leaderboard
8 Gemini 3.8 Flash 100% Sep 2026 Oct 2026 Official leaderboard
9 Claude Opus 5.5 100% Sep 2026 Oct 2026 Official leaderboard
10 GPT-6 Luna 100% Sep 2026 Oct 2026 Official leaderboard
11 GPT-6 Sol 100% Sep 2026 Oct 2026 Official leaderboard
12 Gemini 3.6 Flash 0% Jul 2026 Oct 2026 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard Passed 85% in Oct 2026, by Claude Haiku 4.5

Source: Official Credit Firewall Benchmark leaderboard

Claude Haiku 4.5 on the official leaderboard passed 85% in October 2025

Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Haiku 4.5 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Haiku 4.5

Source: Official Credit Firewall Benchmark leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Safety and alignment

All 636 in Safety and alignment
  • FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
  • StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is. Inkling 98.6%
  • Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
  • Covert misbehavior (Thinking Machines) In use Agentic tasks that check whether a model misbehaves and conceals it. Scored by the rate of covert misbehavior. Inkling 2 Mini 8.30%
  • Destructive actions (Thinking Machines) In use Agentic tasks that check whether a model takes destructive actions it was not authorized to take. Scored by the rate of destructive actions. Inkling 2 Small 4.70%
  • Misbehavior (Thinking Machines) In use Agentic tasks that check whether a model fabricates results, ignores instructions or hides what it did when a task is hard. Scored by the rate of misbehavior. Inkling 2 Mini 9.80%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (12 entries).
  • Pass Models scored 12 models we track.
  • Pass Newest score The newest is from September 2026, for Claude Opus 5.5.
  • Note Frontier models 1 of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Opus 5.5 (#5).
  • Note Headroom Saturated. The best is 100%, past 85% since October 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on kaggle.com: 12 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check