GLM-5.3-Flash
Made by Z.ai (Zhipu AI), released 20 Aug 2026. Epoch AI lists its access as “Open weights (unrestricted)”. Appears strongest at coding among the models listed here.
In short
What it does well
- Coding: Top tier (13th of 77 listed models)
- Writing: Strong (34th of 77 listed models)
Where it falls short
- Maths: Behind the leaders (34th of 43 listed models)
What we don't know yet
- Short factual questions: Not enough data
- Fixing real software bugs: Not enough data
The four ratings
- Top tier
- Strong
- Middle
- Behind the leaders
- Not enough data
Ratings compare only the 77 models listed here, in quarters, best first. “Not enough data” means no result yet. It is never a low score.
- Writing
- StrongHow often people preferred its answers to writing prompts. 1,441 on Arena's scale · Source for GLM-5.3-Flash writing score (lmarena)
- Coding
- Top tierHow often people preferred its answers to coding questions. 1,509 on Arena's scale · Source for GLM-5.3-Flash coding score (lmarena)
- Maths
- Behind the leadersHow many very hard maths problems it solved. 55.8% · Source for GLM-5.3-Flash maths score (epoch)
- Knowledge
- Not enough dataNo result yet for short factual questions.
Evidence completeness: Broad evidence, 11 of 13 tests have a result.
Where it is relatively stronger or weaker
This compares each kind of request with the model's own overall position among the 77 listed models, using Arena's votes. It is about performance on different languages, kinds of request and topics, not about political or social bias.
Relatively stronger
No category stands out as clearly stronger than its overall position.
Relatively weaker
- German: 33rd of 55 on questions asked in German, against 24th of 77 overall. Source for German
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 15th of 77 | 1,514 95% range 1,504 to 1,523 | 4,239 | Source for Expert-level prompts |
| Hard prompts Kind of request | 16th of 77 | 1,492 95% range 1,487 to 1,498 | 26,491 | Source for Hard prompts |
| Long prompts Kind of request | 18th of 77 | 1,484 95% range 1,478 to 1,490 | 18,242 | Source for Long prompts |
| Longer conversations Kind of request | 29th of 77 | 1,471 95% range 1,463 to 1,479 | 6,403 | Source for Longer conversations |
| Chinese Language | 17th of 76 | 1,530 95% range 1,519 to 1,541 | 3,299 | Source for Chinese |
| English prompts Language | 28th of 77 | 1,475 95% range 1,469 to 1,481 | 15,227 | Source for English prompts |
| French Language | 23rd of 72 | 1,488 95% range 1,472 to 1,505 | 1,571 | Source for French |
| German Language | 33rd of 55 | 1,455 95% range 1,432 to 1,478 | 726 | Source for German |
| Japanese Language | 23rd of 54 | 1,450 95% range 1,427 to 1,472 | 834 | Source for Japanese |
| Korean Language | 18th of 54 | 1,442 95% range 1,419 to 1,465 | 741 | Source for Korean |
| Non-English prompts Language | 25th of 77 | 1,463 95% range 1,458 to 1,469 | 26,402 | Source for Non-English prompts |
| Russian Language | 29th of 77 | 1,467 95% range 1,459 to 1,476 | 5,882 | Source for Russian |
| Spanish Language | 19th of 70 | 1,473 95% range 1,457 to 1,489 | 1,574 | Source for Spanish |
| Entertainment, sport and media Topic | 31st of 77 | 1,442 95% range 1,434 to 1,449 | 10,525 | Source for Entertainment, sport and media |
| Law and government Topic | 30th of 76 | 1,475 95% range 1,464 to 1,486 | 3,332 | Source for Law and government |
| Medicine and health Topic | 24th of 75 | 1,478 95% range 1,466 to 1,490 | 2,794 | Source for Medicine and health |
| Science Topic | 25th of 77 | 1,484 95% range 1,476 to 1,492 | 6,813 | Source for Science |
| Software and IT Topic | 14th of 77 | 1,502 95% range 1,495 to 1,508 | 15,570 | Source for Software and IT |
Bias: what we can and can't see
Our sources do not measure political lean, stereotyping, flattery (agreeing with you too readily) or safety behaviour, so we cannot say how any model does on those. What we can show is above: where a model is relatively stronger or weaker across languages, kinds of request and topics. Also, people voting in Arena can favour answer style, such as length and formatting, and our numbers do not adjust for that. See what AI bias means and what these tests can tell you.
Best-supported use cases
- Research and checking facts: not ranked, because there is not enough data for the main test.
Where an alternative might be more suitable
- Writing and creating content: rated Strong on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: rated Behind the leaders on hard maths problems. Compare it with GPT-6.1 Sol, the best-supported choice for this task.
- Business emails and everyday work: rated Strong on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- Research and checking facts: no result on short factual questions, so we cannot say. See which models the evidence does support.
What it is
- Maker
- Z.ai (Zhipu AI)
- Release date (Epoch AI)
- 20 Aug 2026
- Access, as listed by Epoch AI
- Open weights (unrestricted)
- Licence, as listed by Arena
- MIT
- Maker's country (Epoch AI)
- China
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
We have not verified an official product page for this model, so we do not link to one.
How you can use it
Open weights. Arena lists its licence as MIT. In plain terms, the model's files are published under a licence that lets others run it, so it can be offered by more than one company, sometimes at lower cost. That does not guarantee that any particular app is free. Open licences differ, and some restrict commercial use, so check the licence terms before building a business on it.
How you can reach it (an app, a website, a developer API or a third-party tool) and what that costs are not covered by our benchmark data. Where we could verify prices from the maker's own pages, they are shown under Price. Epoch AI lists access as “Open weights (unrestricted)”. See models, apps and harnesses and how AI pricing works.
Price
Price not independently verified
See GLM Coding Plan (checked 28 Oct 2026). Page opened and read. It names GLM models but states no prices or which plan includes which.
We never infer subscription prices from API prices. New to “tokens” and “per million”? See how AI pricing works.
Good to remember
- Benchmarks are samples of tasks. None proves performance in every situation.
- Arena scores come from people choosing between answers. That measures preference, not correctness.
- Ratings are relative to the 77 models listed here. A different set would change them.
- Sources test models at different settings. Each result shows the setting used, and we do not adjust it.
The evidence, for the curious
Show all 13 tests with settings and sources
| Test | Result | Rating | Setting used | Source identifier | Observed | Links |
|---|---|---|---|---|---|---|
| Arena text: overall preference Arena 'text' leaderboard (Bradley-Terry rating, no style control) |
1,473 Arena score 95% range 1,468 to 1,478, 41,642 votes |
Strong | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · text/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,441 Arena score 95% range 1,434 to 1,449, 8,779 votes |
Strong | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · text/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,509 Arena score 95% range 1,502 to 1,516, 10,603 votes |
Top tier | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · text/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,502 Arena score 95% range 1,488 to 1,515, 2,131 votes |
Top tier | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · text/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,481 Arena score 95% range 1,474 to 1,487, 14,398 votes |
Top tier | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · text/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,461 Arena score 95% range 1,453 to 1,469, 7,532 votes |
Strong | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · text/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,607 Arena score 95% range 1,600 to 1,613, 21,587 votes |
Strong | effort: not stated, bradley-terry rating | GLM-5.3-Flash lmarena · webdev/latest |
28 Oct 2026 | Source for GLM-5.3-Flash Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
55.8% share of problems solved standard error ±2.9 points |
Behind the leaders | effort: max, best score across scorers | glm-5.3-flash_max epoch · frontiermath_tiers_1_3_v2.csv |
27 Aug 2026 | Source for GLM-5.3-Flash FrontierMath Tiers 1-3 (v2) score (epoch) No public run log in the data file. |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
93.9% share of problems solved standard error ±2.2 points |
Middle | effort: max, best score across scorers | glm-5.3-flash_max epoch · otis_mock_aime_2024_2025.csv |
26 Aug 2026 | Source for GLM-5.3-Flash OTIS Mock AIME 2024-2025 score (epoch) No public run log in the data file. |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
90.2% share of questions correct standard error ±1.7 points |
Middle | effort: max, best score across scorers | glm-5.3-flash_max epoch · gpqa_diamond.csv |
26 Aug 2026 | Source for GLM-5.3-Flash GPQA Diamond score (epoch) No public run log in the data file. |
| SimpleQA Verified | Not enough data | |||||
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) FrontierCode, benchmark released 2026-06-08, aggregation Mean@5 |
31.8% mean score over 5 attempts no uncertainty figure published |
Not rated (results use different tools) | effort: max, tool: devin-cli, mean@5 | glm-5.3-flash_max epoch · frontiercode_external.csv |
not provided | Source for GLM-5.3-Flash FrontierCode (reported by Cognition) score (epoch) · Run details for FrontierCode (reported by Cognition) |
Missing evidence
Not enough data for these tests:
- SimpleQA Verified
- SWE-bench Verified (Epoch runs)
Missing results are not low scores and do not count against the model.
Where a source gives no public run log, we cannot show the run behind the score. See data sources.
Not sure this is the right one?
Tell us what you need help with and we'll narrow the options down.