GLM 5.3 Flash (v3)
GLM 5.3 Flash (v3) is developed by Zhipu. MeetNewAI lists 5 benchmark boards and 23 plans for it.
Plans and real price
Cheapest measured$0.0029per million tokens, measuredGLM Coding Pro (returning customers ¥149) off-peak · high confidence · Sep 3, 2026
API list price$0.150 / $0.500input / output per million tokens
| Plan | Monthly price | Usable tokens / month | Real price / MTok | Confidence | Data date |
|---|
| GLM Coding Pro (returning customers ¥149) off-peak | $22 | 7.5B | $0.0029 | high | Sep 3, 2026 |
| Droid Max | $200 | 67.4B | $0.0030 | medium | Sep 29, 2026 |
| GLM Coding Pro (returning customers ¥149) midpoint | $22 | 5.6B | $0.0039 | high | Sep 3, 2026 |
| GLM Coding Max (returning customers ¥469) off-peak | $69 | 17.5B | $0.0040 | high | Sep 3, 2026 |
| GLM Coding Max (returning customers ¥469) midpoint | $69 | 13.1B | $0.0053 | high | Sep 3, 2026 |
| GLM Coding Lite (returning customers ¥49) off-peak | $7.23 | 1.2B | $0.0058 | high | Sep 3, 2026 |
| GLM Coding Pro (returning customers ¥149) peak | $22 | 3.7B | $0.0059 | high | Sep 3, 2026 |
| OpenCode Go | $10 | 1.7B | $0.0059 | medium | Sep 30, 2026 |
| GLM Coding Lite (returning customers ¥49) midpoint | $7.23 | 936M | $0.0077 | high | Sep 3, 2026 |
| GLM Coding Max (returning customers ¥469) peak | $69 | 8.7B | $0.0079 | high | Sep 3, 2026 |
| Command Code GOAT | $10 | 1.1B | $0.0088 | medium | Sep 30, 2026 |
| GLM Coding Max (new customers ¥1078) off-peak | $159 | 17.5B | $0.0091 | high | Sep 3, 2026 |
| GLM Coding Pro (new customers ¥538) off-peak | $79 | 7.5B | $0.011 | high | Sep 3, 2026 |
| GLM Coding Lite (returning customers ¥49) peak | $7.23 | 624M | $0.012 | high | Sep 3, 2026 |
| Ollama Max | $100 | 8.5B | $0.012 | medium | Sep 6, 2026 |
| Ollama Pro | $20 | 1.7B | $0.012 | medium | Sep 6, 2026 |
| GLM Coding Max (new customers ¥1078) midpoint | $159 | 13.1B | $0.012 | high | Sep 3, 2026 |
| GLM Coding Lite (new customers ¥118) off-peak | $17 | 1.2B | $0.014 | high | Sep 3, 2026 |
| GLM Coding Pro (new customers ¥538) midpoint | $79 | 5.6B | $0.014 | high | Sep 3, 2026 |
| GLM Coding Max (new customers ¥1078) peak | $159 | 8.7B | $0.018 | high | Sep 3, 2026 |
| GLM Coding Lite (new customers ¥118) midpoint | $17 | 936M | $0.019 | high | Sep 3, 2026 |
| GLM Coding Pro (new customers ¥538) peak | $79 | 3.7B | $0.021 | high | Sep 3, 2026 |
| GLM Coding Lite (new customers ¥118) peak | $17 | 624M | $0.028 | high | Sep 3, 2026 |
Data: real-api-pricing · Sources and licenses
Benchmarks
| Configuration | Harness | Score | Cost per task |
|---|
| GLM-5.3-Flash | — | 41.8 | — |
| Configuration | Harness | Score | Cost per task |
|---|
| Artificial Analysis - GLM-5.3-Flash | Artificial Analysis | 32.8 | — |
| Configuration | Harness | Score | Cost per task |
|---|
| glm-5.3-flash | — | 1615.0 (1607.3–1622.7) | — |
| Configuration | Harness | Score | Cost per task |
|---|
| GLM 5.3 Flash | — | -0.3 (-0.8–0.2) | — |
| Configuration | Harness | Score | Cost per task |
|---|
| mini-swe-agent - glm-5-3-flash (max) | mini-swe-agent | 63.4 (59.0–67.8) | $0.482 |
Data: real-api-pricing · Sources and licenses
Codex Radar runs
| Effort | Harness | Radar IQ | Avg cost (USD) | Avg minutes |
|---|
| low | codex | 70.5 | $0.094 | 24.4 |
| high | codex | 92.3 | $0.325 | 48.7 |
| max | codex | 95.9 | $0.298 | 49.2 |
Data: codexradar · Sources and licenses
