Qwen3 30B A3B Thinking 2507 is cheaper than GLM 5 at $0.071/M vs $0.8/M input tokens.
| Spec | GLM 5 | Qwen3 30B A3B Thinking 2507 |
|---|---|---|
| Input price | $0.8/M tokens | $0.071/M tokens |
| Output price | $2.56/M tokens | $0.285/M tokens |
| Context window | 203K tokens | 262K tokens |
| Released | Feb 2026 | Aug 2025 |
Not enough votes to call it. On the specs, GLM 5 has the edge: bigger model tier, newer, major provider backing.
Qwen3 30B A3B Thinking 2507 costs 9.0x less per token.
Qwen3 30B A3B Thinking 2507 is cheaper on both — 11× input, 9.0× output
Qwen3 30B A3B Thinking 2507 uses 67.9x more emoji
Qwen3 30B A3B Thinking 2507 is cheaper than GLM 5 at $0.071/M vs $0.8/M input tokens.
| Spec | GLM 5 | Qwen3 30B A3B Thinking 2507 |
|---|---|---|
| Input price | $0.8/M tokens | $0.071/M tokens |
| Output price | $2.56/M tokens | $0.285/M tokens |
| Context window | 203K tokens | 262K tokens |
| Released | Feb 2026 | Aug 2025 |
Not enough votes to call it. On the specs, GLM 5 has the edge: bigger model tier, newer, major provider backing.
Qwen3 30B A3B Thinking 2507 costs 9.0x less per token.
Qwen3 30B A3B Thinking 2507 is cheaper on both — 11× input, 9.0× output
Qwen3 30B A3B Thinking 2507 uses 67.9x more emoji