CritPtVerified
19.1%
CritPt composite (AA run)
Models / Zhipu
GLM-5.3
GLM · Released 2026-08-14
Z.ai's August 2026 coding refresh of the GLM-5.2 base. API list price $1.40 / $4.40 per 1M tokens — same slot as GLM-5.2.
Current Z.ai coding-plan and API model for long-horizon agent work. Prefer sourced ledger rows over vendor coding/cyber tables; AA scores sync from the public snapshot. Open weights were promised after a safety delay, not at launch.
Composite54
ContextUnlisted
Input / 1M$1.4
Output / 1M$4.4
Knowledge cutoffunpublished
Output speed66 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
79/100 avg
- Reasoning
- 92
- Coding
- 59
- Agentic
- 84
- Long context
- 80
4 tested benchmark families
Editor's note
Same base as GLM-5.2; gains are post-training. Vendor Terminal Bench / Z.ai Code Bench / cyber figures stay off the ledger until they appear in a mapped source. Editorial depth still needs a desk pass.Benchmark placements
Per-benchmark positions
Where GLM-5.3 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| SciCode | Coding | #6/ 97 | 59 | Artificial Analysis | 2026-09-28 |
| Terminal-Bench-Science | Agentic | #6/ 9 | 8 | Terminal-Bench-Science 0.1 announcement | 2026-08-27 |
| CritPt | Reasoning | #11/ 12 | 19 | Artificial Analysis — CritPt | 2026-09-01 |
| Terminal-Bench | Agentic | #18/ 167 | 84 | Artificial Analysis | 2026-09-28 |
| GPQA Diamond | Reasoning | #29/ 184 | 92 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #34/ 186 | 42 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #45/ 184 | 80 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #92/ 195 | Artificial Analysis | 2026-10-12 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where GLM-5.3 shows up, it's highlighted.
ReasoningGLM-5.3 · 92
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingGLM-5.3 · 59
- 174
- 273
- 370
AgenticGLM-5.3 · 84
- 191
- 289
- 389
Long contextGLM-5.3 · 80
- 189
- 286
- 385
Newest receipts
8 recentLast 120 days
Benchmark rows added to the public ledger for GLM-5.3 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 66 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 79.67% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 91.7% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 42.3% | Artificial Analysis | 24 |
| SciCode | Coding | 59% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 83.9% | Artificial Analysis | 24 |
| CritPt | Reasoning | 19.1% | Artificial Analysis — CritPt | 51 |
| Terminal-Bench-Science | Agentic | 8.1% | Terminal-Bench-Science 0.1 announcement | 56 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where GLM-5.3 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for GLM-5.3. Scores link to the original source; gaps mean no public row exists yet.
8 of 13 catalog benchmarks have a sourced row for GLM-5.3.
Reasoning
Coding
SciCodeVerified
59%
SciCode (AA run)
Agentic
Terminal-BenchVerified
83.9%
Terminal-Bench v2.1 (AA run)
Terminal-Bench-ScienceVerified
8.1%
TB-Science 0.1 resolution (Claude Code)
Long context
AA-LCRVerified
79.67%
AA-LCR (AA run)
Performance
AA output speedVerified
66 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- GLM-5.3 announced on the GLM Coding Plan. Open weights delayed for safety review.
Sources
Where the numbers come from
- Z.ai GLM-5.3 announcement· checked 2026-08-24
- Z.ai pricing· checked 2026-08-24
- Artificial Analysis GLM-5.3· checked 2026-08-24
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is GLM-5.3?
GLM-5.3 is a Zhipu model: Z.ai's August 2026 coding refresh of the GLM-5.2 base. API list price $1.40 / $4.40 per 1M tokens — same slot as GLM-5.2.
What is GLM-5.3 for?
Current Z.ai coding-plan and API model for long-horizon agent work. Prefer sourced ledger rows over vendor coding/cyber tables; AA scores sync from the public snapshot. Open weights were promised after a safety delay, not at launch.
What does GLM-5.3 cost?
GLM-5.3 lists $1.4 input and $4.4 output per 1M tokens. Pricing last checked 2026-08-24.
How large is the GLM-5.3 context window?
GLM-5.3 does not publish a context window on the sources checked for this page.
What is the VerdictPal status of GLM-5.3?
GLM-5.3 carries Solid status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.