Benchmark Almanac

HumanEval-X

CodeGeeX-13B 18.4% tops its leaderboard GPT-NeoX-20B 9.78% leads the models we track
Not in model cards

Models generate programs or translate solutions among Python, C++, Java, JavaScript, and Go. The 820 human-written language variants include executable tests, and scoring uses pass@k.

The code translation task omits the problem description. The star count belongs to the broader CodeGeeX repository.

Made by
THUDM
Released
March 2023
Kind
Function-level problems
Level
School
Scored
Checked automatically
Answers
Code
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Code
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Based on
HumanEval
Last model added
April 2022
Scores refreshed
November 20th, 2026

In the authors’ words

Large pre-trained code generation models, such as OpenAI Codex, can generate syntax- and function-correct code, making the coding of programmers more productive and our pursuit of artificial general intelligence closer. In this paper, we introduce CodeGeeX, a multilingual model with 13 billion parameters for code generation. CodeGeeX is pre-trained on 850 billion tokens of 23 programming languages as of June 2022. Our extensive experiments suggest that CodeGeeX outperforms multilingual code models of similar scale for both the tasks of code generation and translation on HumanEval-X. Building upon HumanEval (Python only), we develop the HumanEval-X benchmark for evaluating multilingual models by hand-writing the solutions in C++, Java, JavaScript, and Go. In addition, we build CodeGeeX-based extensions on Visual Studio Code, JetBrains, and Cloud Studio, generating 4.7 billion tokens for tens of thousands of active users per week. Our user study demonstrates that CodeGeeX can help to increase coding efficiency for 83.4% of its users. Finally, CodeGeeX is publicly accessible and in Sep. 2022, we open-sourced its code, model weights (the version of 850B tokens), API, extensions, and HumanEval-X at https://github.com/THUDM/CodeGeeX.

Abstract of CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-X (arXiv:2303.17568, CC0)
6
Entries in its paper, 2 models we track

CodeGeeX-13B tops the leaderboard with 18.4% GPT-NeoX-20B leads with 9.78%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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 GPT-NeoX-20B, at 9.78%. This view shows only models we track, each at its best.

  1. 1 CodeGeeX-13B Code generation 18.4%
  2. 2 CodeGen-Multi-16B Code generation 16.7%
  3. 3 CodeGen-Multi-6B Code generation 14.3%
  4. 4 InCoder-6.7B Code generation 11.3%
  5. 5 GPT-NeoX-20B Code generation 9.78%
  6. 6 GPT-J-6B Code generation 7.90%

Source: The HumanEval-X paper

Source: The HumanEval-X paper

#EntryScoreSettingDateSource
1 CodeGeeX-13B 18.4% Code generation Official leaderboard
2 CodeGen-Multi-16B 16.7% Code generation Official leaderboard
3 CodeGen-Multi-6B 14.3% Code generation Official leaderboard
4 InCoder-6.7B 11.3% Code generation Official leaderboard
5 GPT-NeoX-20B 9.78% Code generation Apr 2022 Official leaderboard
6 GPT-J-6B 7.90% Code generation Aug 2021 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-NeoX-20B 9.78% Code generation Apr 2022 Official leaderboard
2 GPT-J-6B 7.90% Code generation Aug 2021 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2022 Near the ceiling (85%) GPT-NeoX-20B GPT-J-6B 0% 20% 40% 60% 80% 100% 2022 Near the ceiling (85%) GPT-NeoX-20B GPT-J-6B

Source: Official HumanEval-X 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 Coding

All 523 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 4 years ago, in April 2022, for GPT-NeoX-20B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (6 entries).
  • Pass Models scored 2 models we track, plus 4 other entries on its leaderboard.
  • Warning Newest score The newest came 4 years ago, in April 2022, for GPT-NeoX-20B.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 9.78%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 2 rows for models we track and 4 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