GLM 4.7 Flash vs GLM 5.3 Flash
| Benchmark | GLM 4.7 Flash | GLM 5.3 Flash |
|---|---|---|
| Humanity's Last Exam | 7.6 | 39.9 |
| Terminal-Bench 4.0 | — | 32.8 |
| AutomationBench-AA | — | 60.4 |
Specs
| Spec | GLM 4.7 Flash | GLM 5.3 Flash |
|---|---|---|
| Input, per 1M | !$0.061 | !$0.15 |
| Output, per 1M | !$0.40 | !$0.50 |
| Cost per task? | — | $0.28 |
| Speed, tok/s | 87 | 73 |
| Context | 200K | 1M |
| Weights | Open | Open |
| Released | Jan 2026 | Aug 2026 |
| Memory (4-bit) | 19 GB | 193 GB |
| Licence | MIT | MIT |
| Commercial use | Allowed | Allowed |
| See images | — | ✓ |
| Read PDFs | — | Partial |
| Hear audio | — | — |
| Watch video | — | ✓ |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | — |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
GLM 4.7 Flash → GLM 5.3 Flash
Memory (4-bit): 19 GB → 193 GB
Defaults to max · Self-hosted›
The Flash chat template falls back to max reasoning effort for missing or unrecognized values; pass low or high explicitly to reduce effort.
Reasoning cannot be disabled›
Disabling reasoning is unsupported; keep reasoning enabled.
tool_choice auto · Z.ai API›
The first-party chat reference exposes only auto for tool_choice.
Complete unmodified reasoning_content›
Preserved thinking defaults on for Coding Plan and off for the standard API; when enabled, return complete unmodified reasoning_content in its original order to protect quality and cache hits.
Only one stop word›
Only one stop word is supported, so do not rely on a list of alternative stop sequences.
GLM 5.3 Flash → GLM 4.7 Flash