GLM 5.3
Z.ai · released Aug 18, 2026
GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...
Specification
- Context window
- 1.05M
- Max output
- 944K
- Knowledge cutoff
- Not stated
- Parameters
- 753.3B
- Licence
- other
- Serving providers
- 34
- Moderated
- No
- Uptime
- 100.0%
Intelligence
44.8
95th percentile
Coding
74.8
Coding Index
Agentic
53.1
Agentic Index
Output speed
138 t/s
Median across providers
Latency
714ms
Time to first token
Cost per task
$0.39
Estimated
Benchmarks
Where the score comes from
The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.
Evaluation scores
Percentage correct · higher is better
- GPQA Diamond91.7%
- AA-LCR (long context)79.7%
- SciCode59.0%
- τ²-bench (Telecom)50.3%
- Humanity's Last Exam42.3%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| GPQA Diamond | 91.7% |
| AA-LCR (long context) | 79.7% |
| SciCode | 59.0% |
| τ²-bench (Telecom) | 50.3% |
| Humanity's Last Exam | 42.3% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- Grok 4.7xAI46.4
- MiMo-V2.6-ProXiaomi46.3
- 45.4
- GLM 5.3Z.ai44.8
- Grok 4.6xAI44.3
- Step 5 PreviewStepFun43.7
- Kimi K3Moonshot AI43.6
- Claude Haiku 5.5Anthropic43.4
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| Grok 4.7 | 46.4 |
| MiMo-V2.6-Pro | 46.3 |
| Qwen3.8 Max (0902) | 45.4 |
| GLM 5.3 | 44.8 |
| Grok 4.6 | 44.3 |
| Step 5 Preview | 43.7 |
| Kimi K3 | 43.6 |
| Claude Haiku 5.5 | 43.4 |
Percentile among all indexed models
Pricing
What it costs to run
List prices per million tokens, plus what one representative task works out to.
List price
- Input / 1M tokens
- $0.03
- Output / 1M tokens
- $4.8
- Cached input / 1M
- $0.029
- Blended 3:1
- $1.22
One task, estimated
$0.39
- Input tokens
- 50,000
- Output tokens
- 80,000
- Profile
- Reasoning
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).
Serving providers
- Relace
- Wafer
- InferenceNet
- Morph
- Reka
- Makora
- Sail Research
- AkashML
- Io Net
- Inceptron
- DeepInfra
- Tenstorrent
- SiliconFlow
- Novita
- Phala
- DigitalOcean
- Decart
- GMICloud
- Alibaba
- Friendli
- Mistral
- Baidu
- BaseTen
- Crusoe
- PrimeIntellect
- Venice
- Together
- Parasail
- Modal
- Fireworks
- Cloudflare
- AtlasCloud
- Z.AI
- ModelRun
Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.