Benchmark Almanac

PII Masking

ichetandhembre/Qwen3-4B-Instruct-2507-PII-SFT-LoRA 1.48 tops its leaderboard Claude Sonnet 4.5 0.73 leads the models we track
Not in model cards

Models locate personal information in US English text and replace it with a common redaction tag. The default split reserves 20% for evaluation.

The hub reports Qwen3-4B-Instruct-2507-PII-SFT-LoRA, Qwen3-4B-Instruct-2507-PII-RL, gpt-5-nano and other models. Qwen3-4B-Instruct-2507-PII-SFT-LoRA reached a mean reward of 1.4775 on 200 examples on July 3rd, 2026.

Made by
Adam Lucek
Released
October 2025
Kind
Following precise instructions
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Risk
Privacy
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
In views
Content safety
Strongest model tested
GPT-5 #71 on the capability index
Last model added
September 2025
Scores refreshed
November 20th, 2026
7
Entries on its own leaderboard, 4 models we track

ichetandhembre/Qwen3-4B-Instruct-2507-PII-SFT-LoRA tops the leaderboard with 1.48 Claude Sonnet 4.5 leads with 0.73 among models released by September 2025

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Claude Sonnet 4.5, at 0.73. This view shows only models we track, each at its best.

The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 ichetandhembre/Qwen3-4B-Instruct-2507-PII-SFT-LoRA 0.1.1, 200 examples 1.48
  2. 2 ichetandhembre/Qwen3-4B-Instruct-2507-PII-RL 0.1.1, 50 examples 1.19
  3. 3 AdamLucek/Qwen3-4B-Instruct-2507-PII-RL 0.1.1, 50 examples 0.89
  4. 4 Claude Sonnet 4.5 0.1.1, 50 examples 0.73
  5. 5 GPT-5 0.1.1, 50 examples 0.68
  6. 6 qwen3-4b-instruct-2507 0.1.1, 50 examples 0.63
  7. 7 GPT-5 nano 0.1.1, 50 examples 0.58

Source: Official PII Masking leaderboard

#EntryScoreSettingSubmitted byReleasedRun onLinkSource
1 ichetandhembre/Qwen3-4B-Instruct-2507-PII-SFT-LoRA 1.48 0.1.1, 200 examples, verifiers Chetan Dhembre Jul 2026 app.primeintellect.ai Official leaderboard
2 ichetandhembre/Qwen3-4B-Instruct-2507-PII-RL 1.19 0.1.1, 50 examples, verifiers Chetan Dhembre May 2026 app.primeintellect.ai Official leaderboard
3 AdamLucek/Qwen3-4B-Instruct-2507-PII-RL 0.89 0.1.1, 50 examples, verifiers Adam Lucek Nov 2025 app.primeintellect.ai Official leaderboard
4 Claude Sonnet 4.5 0.73 0.1.1, 50 examples, verifiers Adam Lucek Sep 2025 Nov 2025 app.primeintellect.ai Official leaderboard
5 GPT-5 0.68 0.1.1, 50 examples, verifiers Adam Lucek Aug 2025 Nov 2025 app.primeintellect.ai Official leaderboard
6 qwen3-4b-instruct-2507 0.63 0.1.1, 50 examples, verifiers Adam Lucek Aug 2025 Nov 2025 app.primeintellect.ai Official leaderboard
7 GPT-5 nano 0.58 0.1.1, 50 examples, verifiers Adam Lucek Aug 2025 Nov 2025 app.primeintellect.ai Official leaderboard
#ModelScoreSettingSubmitted byReleasedRun onLinkSource
1 Claude Sonnet 4.5 0.73 0.1.1, 50 examples, verifiers Adam Lucek Sep 2025 Nov 2025 app.primeintellect.ai Official leaderboard
2 GPT-5 0.68 0.1.1, 50 examples, verifiers Adam Lucek Aug 2025 Nov 2025 app.primeintellect.ai Official leaderboard
3 qwen3-4b-instruct-2507 0.63 0.1.1, 50 examples, verifiers Adam Lucek Aug 2025 Nov 2025 app.primeintellect.ai Official leaderboard
4 GPT-5 nano 0.58 0.1.1, 50 examples, verifiers Adam Lucek Aug 2025 Nov 2025 app.primeintellect.ai 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 Not tested since Oct 2025

Source: Official PII Masking leaderboard

Scores over time

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
0.55 0.60 0.65 0.70 0.75 0.55 0.60 0.65 0.70 0.75

Source: Official PII Masking 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 Instructions and chat

All 144 in Instructions and chat
  • IFEval In use Models answer about 500 prompts containing verifiable instructions such as word limits or required keywords. Code checks each instruction for strict and loose accuracy. Gemma 4 31B IT 98.9%
  • IFBench In use 58 new out-of-domain verifiable output constraints. Scored by programmatic constraint checks (strict or loose). Flint 89.2%
  • MultiChallenge In use Multi-turn conversations testing instruction retention, inference memory, versioned editing and self-coherence. Graded by LLM judge with rubrics. GPT-6.1 Sol 82%
  • Multi-IF In use Models follow verifiable instructions over three-turn conversations in eight languages, extending IFEval. Scored by instruction accuracy at each turn. LFM3-1.7B 89.1%
  • AdvancedIF In use Models respond to over 1,600 prompts with complex, multi-turn, and system-level instructions. A judge model checks each response against expert-written rubrics. MAI-Thinking-1 85%
  • MM-IFEval In use Multimodal models follow image-grounded instructions with exact output constraints. Checks mix rule-based verification and a judge model. GPT-4o 64.6%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 14 months ago, in September 2025, for Claude Sonnet 4.5. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (7 entries).
  • Pass Models scored 4 models we track, plus 3 other entries on its leaderboard.
  • Warning Newest score The newest came 14 months ago, in September 2025, for Claude Sonnet 4.5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on app.primeintellect.ai: 4 rows for models we track and 3 other entries 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