Baseten
99.86%Context1M
Max output1M
p50 latency1,514 ms
Throughput138.0 t/s
Input price$1.4
Output price$4.4
Zero retentionNo trainingCaching
Model directory
Baseer model profileLanguage model
zai/glm-5.3
GLM 5.3
GLM 5.3 delivers comprehensive advancements in complex software engineering and agent capabilities. It uses the same base model as GLM-5.2, with all improvements driven by post-training.
Input / 1M tokens$0.7
Output / 1M tokens$2.2
Maximum context1M
Inference providers18
Price comparison
USD per one million tokens, using current provider data.
Provider latency
Last-hour p50 in milliseconds; lower is better.
Provider uptime
Reported availability over the last 24 hours; not a contractual SLA.
Inference providers
Compare current pricing, performance, and privacy across every available provider.
Blackbox
55.88%Context1M
Max output262.1K
p50 latency—
Throughput—
Input price$0.7
Output price$2.2
Caching
Crusoe
91.28%Context1M
Max output1M
p50 latency5,081 ms
Throughput415.0 t/s
Input price$1.4
Output price$4.4
Caching
Deepinfra
100.00%Context1M
Max output1M
p50 latency386 ms
Throughput144.0 t/s
Input price$1.2
Output price$4
Zero retentionNo trainingCaching
Digitalocean
99.96%Context1M
Max output1M
p50 latency1,388 ms
Throughput200.0 t/s
Input price$1.4
Output price$4.4
Zero retentionNo trainingCaching
Fireworks
99.99%Context1M
Max output1M
p50 latency1,427 ms
Throughput103.0 t/s
Input price$1.4
Output price$4.4
Zero retentionNo trainingCaching
Friendli
100.00%Context1M
Max output1M
p50 latency153 ms
Throughput221.5 t/s
Input price$1.26
Output price$3.96
Caching
Gmicloud
98.97%Context1M
Max output1M
p50 latency902 ms
Throughput126.0 t/s
Input price$1.12
Output price$3.52
Caching
Inceptron
99.86%Context1M
Max output1M
p50 latency666 ms
Throughput43.0 t/s
Input price$1.25
Output price$4.4
No trainingCaching
Modal
99.81%Context1M
Max output1M
p50 latency452 ms
Throughput157.5 t/s
Input price$1.4
Output price$4.4
Zero retentionNo trainingCaching
Morph
100.00%Context1M
Max output1M
p50 latency788 ms
Throughput208.5 t/s
Input price$1.25
Output price$4.4
Zero retentionNo trainingCaching
Novita
100.00%Context1M
Max output128K
p50 latency677 ms
Throughput73.5 t/s
Input price$1.4
Output price$4.4
No trainingCaching
Parasail
100.00%Context1M
Max output1M
p50 latency501 ms
Throughput93.5 t/s
Input price$1.4
Output price$4.4
Zero retentionNo trainingCaching
Runware
100.00%Context1M
Max output1M
p50 latency2,174 ms
Throughput102.5 t/s
Input price$1.2
Output price$4
Caching
Streamlake
100.00%Context1M
Max output128K
p50 latency1,380 ms
Throughput84.0 t/s
Input price$1.4
Output price$4.4
Caching
Togetherai
100.00%Context1M
Max output1M
p50 latency121 ms
Throughput424.0 t/s
Input price$1.4
Output price$4.4
Zero retentionNo trainingCaching
Wafer
100.00%Context1M
Max output1M
p50 latency346 ms
Throughput—
Input price$1.19
Output price$4.4
Zero retentionNo trainingCaching
Zai
99.85%Context1M
Max output128K
p50 latency4,172 ms
Throughput31.0 t/s
Input price$1.4
Output price$4.4
Caching
| Provider | Maximum context | Max output | p50 latency | p50 throughput | 24h uptime | Input / 1M tokens | Output / 1M tokens | Privacy |
|---|---|---|---|---|---|---|---|---|
Baseten | 1M | 1M | 1,514 ms | 138.0 t/s | 99.862% | $1.4 | $4.4 | |
Blackbox | 1M | 262.1K | — | — | 55.880% | $0.7 | $2.2 | |
Crusoe | 1M | 1M | 5,081 ms | 415.0 t/s | 91.279% | $1.4 | $4.4 | |
Deepinfra | 1M | 1M | 386 ms | 144.0 t/s | 100.000% | $1.2 | $4 | |
Digitalocean | 1M | 1M | 1,388 ms | 200.0 t/s | 99.957% | $1.4 | $4.4 | |
Fireworks | 1M | 1M | 1,427 ms | 103.0 t/s | 99.992% | $1.4 | $4.4 | |
Friendli | 1M | 1M | 153 ms | 221.5 t/s | 100.000% | $1.26 | $3.96 | |
Gmicloud | 1M | 1M | 902 ms | 126.0 t/s | 98.974% | $1.12 | $3.52 | |
Inceptron | 1M | 1M | 666 ms | 43.0 t/s | 99.860% | $1.25 | $4.4 | |
Modal | 1M | 1M | 452 ms | 157.5 t/s | 99.815% | $1.4 | $4.4 | |
Morph | 1M | 1M | 788 ms | 208.5 t/s | 100.000% | $1.25 | $4.4 | |
Novita | 1M | 128K | 677 ms | 73.5 t/s | 100.000% | $1.4 | $4.4 | |
Parasail | 1M | 1M | 501 ms | 93.5 t/s | 100.000% | $1.4 | $4.4 | |
Runware | 1M | 1M | 2,174 ms | 102.5 t/s | 100.000% | $1.2 | $4 | |
Streamlake | 1M | 128K | 1,380 ms | 84.0 t/s | 100.000% | $1.4 | $4.4 | |
Togetherai | 1M | 1M | 121 ms | 424.0 t/s | 100.000% | $1.4 | $4.4 | |
Wafer | 1M | 1M | 346 ms | — | 100.000% | $1.19 | $4.4 | |
Zai | 1M | 128K | 4,172 ms | 31.0 t/s | 99.851% | $1.4 | $4.4 |
What can this model do?
Available modalities and capabilities across inference providers.
TextAdvanced reasoningTool useAutomatic cachingTools
Reasoning options
LowHighMax
Technical profile
- Model type
- Language model
- Release date
- Aug 18, 2026
- Knowledge cutoff
- —
- Regions
- US
- API specifications
- V2, V3, V4
- Temperature control
- Supported
Model identifier
zai/glm-5.3Model profile updated: Sep 12, 2026, 1:46 PM
Related models
More options from the same creator or model category.