GLM 5.3 (batch) is Z.ai's reasoning model optimized for asynchronous batch processing with a 1M-token context window.
- Strengths
- Handles large batches of requests efficiently, suited for non-time-sensitive workloads where throughput and resource efficiency matter more than latency.
- Best for
- Processing large document collections, log analysis, content generation pipelines, and other batch jobs where requests can be queued and processed asynchronously.
- Limitations
- Designed for batch processing rather than real-time interactions; not suitable for applications requiring immediate responses or interactive conversation.
Input / 1M
$0.45
Output / 1M
$2.00
Cached input / 1M
$0.1
Context window
1.05M
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in — out —
- 90d
- in decreased 35.7% out decreased 9.1%
- 1y
- in decreased 35.7% out decreased 9.1%
- Since launch
- in decreased 35.7% out decreased 9.1%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 23 Sep 2026 | $0.45 | $2.00 | $0.1 | Imported from OpenRouter | openrouter.ai |
| 22 Sep 2026 | $0.72 | $2.40 | $0.12 | Imported from OpenRouter | openrouter.ai |
| 8 Sep 2026 | $0.7 | $2.20 | $0.13 | Imported from OpenRouter | openrouter.ai |