Model Scorecard
GLM-5.3 (max)
Zhipu (GLM) · open weights
GLM-5.3 ranks #9 of 81 models on the October 2026 ToneBench board, and GLM-5.3 (max) is its only configuration, #17 of 134 configurations, with a Writing Score of 82.4/100 and a Writing Elo of 2194. One script costs $0.240 at API list price.
Last updated , from board data generated at 16:59 UTC.
Rank, scores, the nine metrics, the 10 reference scripts, cost and setup for one configuration (a model at one reasoning setting). Key Results sums it up; the section links below it jump to the detail.
Key Results
- StandingGLM-5.3 ranks #9 of 81 models, and this is its only configuration on the board. As a configuration, it is #17 of 134 configurations by Writing Elo (2194), in the top 20. Its 95% range (2161 to 2230) overlaps 11 other configurations, ranked #12 to #23, so its exact place among them is not settled.
- MetricsBest-ranked: Hook Strength (83.4, 4th of 134 configurations, +4.2 against the board median). Lowest-ranked: Substance, Accuracy & Value (83.2, 45th of 134 configurations, +2.4 against the board median).
- ScriptsFrom 80.5 on Script 10 (engineering process walkthrough) to 84.1 on Script 7 (personal technical walkthrough): a 3.6-point spread across the 10 reference scripts.
- ConsistencyIts Writing Score moves ± 1.4 between drafts of the same script, against ± 2.8 for the median configuration: steadier than most.
- ValueGPT-6 Sol (medium) ranks about the same (their Elo ranges overlap) and costs about $0.056 per script, against $0.240 per script for this one.
Louis's PickI'd pick GPT-6 Sol (max) over this one: it ranks clearly higher for no more per script.
More Context Behind These Results
Writing Elo comes from comparing its average on each script with every other configuration's average on that same script; a gap smaller than the draft-to-draft variation counts as a draw (what the ranges can and cannot show).
On Hook Strength, its best-ranked metric, the judges rated its openings above the typical configuration's. Even its lowest-ranked metric, Substance, Accuracy & Value, ranks in the top half of the board, so no metric stands out as a weak spot.
The script format moves its score. If most of your scripts look like its best or weakest format, that result is a better guide than the average.
Its drafts vary little, so the first draft is usually close to the next one.
It outranks 9 configurations that cost at least 50% more.
It ranks highest of the 7 Zhipu (GLM) configurations on the board, though its Elo range overlaps that of GLM-5.3 Flash (max).
It is the highest-ranked open-weight configuration on the board, 1st of 43, with only proprietary configurations above it, so it is the first one to try when the weights have to run on your own hardware.
Skill Profile Across the Nine Metrics
Each corner is one metric on a 0 to 100 scale; the further out, the higher the score. A corner outside the dashed line beats the board median. Hover or tap a point for its exact score, and a metric name for what it rewards.
Scores by Metric
Nine metrics, each scored 0 to 100 and weighted into the Writing Score; the weight sits under each name. Rank is out of all 134 ranked configurations.
Scores by Reference Script
Every ranked configuration writes five drafts for each of the same 10 briefs, and each score is the mean of those drafts. The rank and board median compare it with every configuration on that script, which separates a weak result from a hard script. The result on the format closest to your own work is the most relevant.
Consistency Between Drafts
Every score on this page is the mean of 5 drafts per script. The ± is the standard deviation between those drafts: the larger it is, the more two drafts of the same script can differ.
Compared with the typical model, it's steadiest on Length Adherence: ± 3.3 between drafts against ± 7.0.
Consistency by Metric
| Metric | This model | Board median | Verdict |
|---|---|---|---|
| Tone & Voice Match ? | 79.7 ± 1.1 | ± 2.0 | steadier than most |
| Writing Craft & Clarity ? | 78.7 ± 1.2 | ± 1.6 | typical |
| Substance, Accuracy & Value ? | 83.2 ± 2.1 | ± 2.3 | typical |
| Continuity & Emotion ? | 82.1 ± 1.2 | ± 1.5 | typical |
| YouTube Best Practices ? | 81.7 ± 1.6 | ± 2.5 | typical |
| Hook Strength ? | 83.4 ± 2.2 | ± 2.8 | typical |
| Length Adherence ? | 94.3 ± 3.3 | ± 7.0 | steadier than most |
| Slop Score ? | 83.4 ± 1.3 | ± 1.6 | typical |
| Visual Cue Quality ? | 79.7 ± 2.3 | ± 3.0 | typical |
Cost, Time and Tokens per Script
Logged on every run; none of these figures affects the Writing Score. Time is the average time to one accepted script, retries included. Cost is one script at the API list price.
Scores by Judge
Four judges score the same 50 stored drafts blind against the same rubric: three LLM judges from three labs, plus Jev, a typed judge from TypeSafe that rates each metric on five described quality levels. No single judge's taste decides the result. The Writing Score of 82.4 combines their scores with the parts computed in code (Length Adherence and half of the Slop Score). The highest and lowest judge are 12.7 points apart on its Writing Score. That is close to the board median of 13.7. Each judge also has its own scale, so every card shows that judge's score against its average across the board. How the panel works.
Evaluation Setup
How this exact configuration was run and judged: route, model ID, dates, sampling and prompt.
- Accessed viaOpenRouter · model page ↗.
- Exact model / route id
z-ai/glm-5.3(the request sets no level; the provider applies max). - Generations5 independent runs × 10 scripts = 50 drafts, stored after generation and never edited.
- Run dates2026-10-08.
- JudgesClaude Opus 5.5 at medium effort through the Claude Code subscription, GPT-6.1 Sol at medium reasoning through the Codex CLI, GLM 5.3 Flash through OpenRouter, and Jev 1.13 from TypeSafe. One call per judge per draft. All four score every draft on their own, blind to which model wrote it. The three LLM judges use the identical rubric and have to quote evidence before they score a metric. Jev answers the same rubric as one question per metric, choosing among five described quality levels, and writes no text. Seven metrics are the plain average of the four judges. Length is computed from the spoken word count, and Slop is half computed (lexical scans minus the spoken-dash penalty) and half judged. The Writing Elo is cross-checked with a Bradley-Terry fit of the same pairwise results (methodology).
- Sampling and promptThe provider's default sampling, with no fine-tuning and no system-prompt changes. Every current-ranked model gets the same prompt, about 13,000 tokens (the count shifts a bit with each model's tokenizer): a style guide built from Louis's voice and storytelling guides, the target length, and the brief with its research packet. Generation is closed-book on every route: the prompt is the only source, with no web search, browsing, code execution, file access or other tool (generation).
- What stays privateThe exact rubric and any non-public reference material. Each script is normally benchmarked before its video is published, which lowers the odds that a model has seen the reference, though it does not rule it out (why).
Frequently Asked Questions
How Good Is GLM-5.3 (max) at Writing?
GLM-5.3 ranks #9 of 81 models on ToneBench, and this is its only configuration on the board. This configuration is #17 of 134 configurations, in the top 20, with a Writing Elo of 2194 and a Writing Score of 82.4/100. The ranks come from full YouTube scripts written in the voice of Louis-François Bouchard and judged blind, with the channel's finished script as the reference. The benchmark measures long-form scripted writing, so the result does not carry over to every kind of text.
Is GLM-5.3 (max) the Best Zhipu (GLM) Model for Writing?
Yes. Of the 7 Zhipu (GLM) configurations on the board, GLM-5.3 (max) ranks highest by Writing Elo, which makes GLM-5.3 the top Zhipu (GLM) model, #9 of 81 models. Its Elo range overlaps that of GLM-5.3 Flash (max), so the order between those two is not settled.
Is GLM-5.3 (max) Good Value for the Money?
It costs about $0.240 per script. GPT-6 Sol (medium) ranks about the same, inside the overlap of their Elo ranges, and costs less.
What Are the Strengths and Weaknesses of GLM-5.3 (max)?
Its best-ranked metric is Hook Strength (83.4, 4th of 134 configurations, +4.2 against the board median), and its lowest-ranked is Substance, Accuracy & Value (83.2, 45th of 134, +2.4). Even that one ranks in the top half of the board. The Scores by Metric section shows the board median and the board's best for each one.
How Was GLM-5.3 (max) Evaluated?
It was run through OpenRouter, with the exact model/route ID z-ai/glm-5.3, first run on 2026-10-08. It wrote each script 5 times with the provider's default sampling and no fine-tuning. The Evaluation Setup section and the methodology page have the details.
Does GLM-5.3 (max) Have Open Weights?
Yes. Its weights are public, so it can be downloaded and run on your own hardware. Check the license if you need open source in the strict sense. Among open-weight configurations it is 1st of 43 for writing.