Benchmark Almanac

Bug in the Code Stack

GPT-4o Mar 2025 82.7% leads since March 2025
Not in model cards

Models find a syntax error inserted into a long Python codebase. Hamming measures retrieval accuracy across context lengths and error positions.

Hamming lists results for 11 models.

Made by
Hamming AI and Wat.ai
Kind
Finding facts in long inputs
Level
School
Scored
Checked automatically
Answers
Short answer
Measures
Capability
Language
English
In and out
Text, Code → Text
Interaction
Single turn
Refresh
One fixed set
Run by
A host
Strongest model tested
Gemini 1.5 Pro Sep 2024 #157 on the capability index
Last model added
March 2025
Scores refreshed
November 20th, 2026
11
Entries on its own leaderboard, 9 models we track

GPT-4o leads with 82.7% among models released by March 2025 GPT-4o leads with 82.7% among models released by March 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. This view shows only models we track, each at its best.

The newest model with a score here came out in March 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 GPT-4o Mar 2025 82.7%
  2. 2 GPT-4 Turbo Apr 2024 78.1%
  3. 3 Claude 3 Opus 59.5%
  4. 4 Gemini 1.5 Pro Sep 2024 52.5%
  5. 5 Gemini 1.5 Flash Sep 2024 49.4%
  6. 6 GPT-3.5 Turbo Jan 2024 36.5%
  7. 7 Codestral 22B 36.3%
  8. 8 Llama 3-70B 35.4%
  9. 9 Command R+ 30.1%
  10. 10 Gemini 1.0 Pro 16.3%
  11. 11 CodeQwen 1.5 0.48%

Source: Official Bug in the Code Stack leaderboard

#EntryScoreSettingDateSource
1 GPT-4o Mar 2025 82.7% Mar 2025 Official leaderboard
2 GPT-4 Turbo Apr 2024 78.1% Apr 2024 Official leaderboard
3 Claude 3 Opus 59.5% Feb 2024 Official leaderboard
4 Gemini 1.5 Pro Sep 2024 52.5% Sep 2024 Official leaderboard
5 Gemini 1.5 Flash Sep 2024 49.4% Aug 2024 Official leaderboard
6 GPT-3.5 Turbo Jan 2024 36.5% Jan 2024 Official leaderboard
7 Codestral 22B 36.3% Official leaderboard
8 Llama 3-70B 35.4% Apr 2024 Official leaderboard
9 Command R+ 30.1% Aug 2024 Official leaderboard
10 Gemini 1.0 Pro 16.3% Dec 2023 Official leaderboard
11 CodeQwen 1.5 0.48% Official leaderboard
#ModelScoreSettingDateSource
1 GPT-4o Mar 2025 82.7% Mar 2025 Official leaderboard
2 GPT-4 Turbo Apr 2024 78.1% Apr 2024 Official leaderboard
3 Claude 3 Opus 59.5% Feb 2024 Official leaderboard
4 Gemini 1.5 Pro Sep 2024 52.5% Sep 2024 Official leaderboard
5 Gemini 1.5 Flash Sep 2024 49.4% Aug 2024 Official leaderboard
6 GPT-3.5 Turbo Jan 2024 36.5% Jan 2024 Official leaderboard
7 Llama 3-70B 35.4% Apr 2024 Official leaderboard
8 Command R+ 30.1% Aug 2024 Official leaderboard
9 Gemini 1.0 Pro 16.3% Dec 2023 Official leaderboard

5 models have held number one on the official leaderboard since 2023

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

Google DeepMind OpenAI Anthropic
Own leaderboard Not tested since Mar 2025
GPT-4o 82.7%

Source: Official Bug in the Code Stack 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 Best so far
0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) GPT-4o GPT-4 Turbo Claude 3 Opus GPT-3.5 Turbo Gemini 1.0 Pro 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) GPT-4o Claude 3 Opus

Source: Official Bug in the Code Stack 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 Long context

All 115 in Long context
  • MRCR v2 In use Multi-round coreference: retrieve the n-th of several similar requests hidden in a long synthetic conversation, up to 1M tokens. Muse Spark 1.3 98.5%
  • LongBench v2 In use 503 multiple-choice questions over 8K-2M-word contexts (QA, dialogue, code repos, structured data). Accuracy. Inkling 2 Small 70.6%
  • GraphWalks In use A graph is given as an edge list in long context. The model answers BFS or parent queries, scored by F1. Claude Fable 5 99.96%
  • OpenAI MRCR v2 In use Models read a long synthetic chat that hides 2, 4 or 8 identical writing requests and must reproduce the response to the i-th one. Scored by mean string match ratio per context bin. Gemini 3.7 Flash 89%
  • CL-bench In use Model must learn new knowledge from a supplied long context (rules, procedures, data) and solve tasks. Strict LLM-verified rubrics. GPT-5.4 27.9%
  • AA-LCR 1.1 In use Models answer 100 questions that need reasoning across sets of real documents of about 100K tokens. An LLM equality checker with a fixed system prompt compares each answer with the key. Step 5 Preview 88.3%

Data audit

8 of 10 checks pass, 2 warnings. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (11 entries).
  • Pass Models scored 9 models we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 20 months ago, in March 2025, for GPT-4o.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 1.5 Pro (#157).
  • Pass Headroom The best is 82.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on hamming.ai: 9 rows for models we track and 2 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