Benchmark Almanac

WordEditBench

Tencent Hy4 57.2 leads since November 2026
Not in model cards

Agents edit Word documents. A rule-based checker reads the document XML, rewards requested changes, and penalizes changes to unrelated content.

Tencent reports seven models on 100 tasks in the Hy4 flash appendix. Tasks include applying formatting from another document and formatting a thesis.

Made by
Tencent
Released
November 2026
Kind
Documents and spreadsheets
Level
University
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Measures
Capability
In and out
Text, Documents → Documents, Actions
Interaction
Agentic
Environment
Microsoft Word
Refresh
One fixed set
Run by
A host
In views
Agents
Strongest model tested
DeepSeek V4.1 Flash 1017 #23 on the capability index
Last model added
November 2026
Scores refreshed
November 21st, 2026
7
Entries on its own leaderboard, 6 models we track

Tencent Hy4 leads with 57.2 Tencent Hy4 leads with 57.2

Best score per entry from the benchmark's own leaderboard, read on November 21st, 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.

  1. 1 Tencent Hy4 highest available reasoning 57.2
  2. 2 DeepSeek V4.1 Flash 1017 highest available reasoning 53.5
  3. 3 GLM-5.3-Flash highest available reasoning 47.6
  4. 4 Qwen3.8-Flash-Next highest available reasoning 46.3
  5. 5 Hy4 flash highest available reasoning 46.2
  6. 6 Tencent Hy4 preview highest available reasoning 44.2
  7. 7 GPT-6 Luna highest available reasoning 42.7

Source: Official WordEditBench leaderboard

#EntryScoreSettingDateSource
1 Tencent Hy4 57.2 highest available reasoning Nov 2026 Official leaderboard
2 DeepSeek V4.1 Flash 1017 53.5 highest available reasoning Oct 2026 Official leaderboard
3 GLM-5.3-Flash 47.6 highest available reasoning Aug 2026 Official leaderboard
4 Qwen3.8-Flash-Next 46.3 highest available reasoning Aug 2026 Official leaderboard
5 Hy4 flash 46.2 highest available reasoning Official leaderboard
6 Tencent Hy4 preview 44.2 highest available reasoning Aug 2026 Official leaderboard
7 GPT-6 Luna 42.7 highest available reasoning Sep 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Tencent Hy4 57.2 highest available reasoning Nov 2026 Official leaderboard
2 DeepSeek V4.1 Flash 1017 53.5 highest available reasoning Oct 2026 Official leaderboard
3 GLM-5.3-Flash 47.6 highest available reasoning Aug 2026 Official leaderboard
4 Qwen3.8-Flash-Next 46.3 highest available reasoning Aug 2026 Official leaderboard
5 Tencent Hy4 preview 44.2 highest available reasoning Aug 2026 Official leaderboard
6 GPT-6 Luna 42.7 highest available reasoning Sep 2026 Official leaderboard

2 models have held number one on the official leaderboard since its release in November 2026

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

DeepSeek Tencent
Own leaderboard 2 leaders

Source: Official WordEditBench 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
40.0 45.0 50.0 55.0 60.0 2027 40.0 45.0 50.0 55.0 60.0 2027

Source: Official WordEditBench 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 Economic work

All 210 in Economic work
  • JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
  • GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate. Seed2.1 Pro 87.9%
  • GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
  • AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
  • Agents' Last Exam v1.1 In use Agents complete 122 public long-horizon professional workflow tasks across 55 sub-industries. Deterministic graders check outcomes against hidden references. GPT-6.2 Astra 47.5%
  • CoWorkBench In use Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished). Claude Fable 5 75.9%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (7 entries).
  • Pass Models scored 6 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Tencent Hy4.
  • Note Frontier models None of the top 10 has a score. The strongest tested is DeepSeek V4.1 Flash 1017 (#23).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on huggingface.co: 6 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 21st, 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