Gemini 3.7 Flash vs GLM-4.5
Gemini 3.7 Flash (Google DeepMind) and GLM-4.5 (Z.ai (Zhipu AI),Tsinghua University) compared on benchmarks, pricing, context window, and use-case rankings.
- GLM-4.5 is cheaper on input tokens — $0.60 vs $0.75 per 1M.
- Gemini 3.7 Flash has a larger context window (1M vs 131K tokens).
- Gemini 3.7 Flash ranks higher for coding (88th vs 19th percentile).
| Attribute | Gemini 3.7 Flash | GLM-4.5 |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 157 | — |
| Coding | 88 | 19 |
| Math | 74 | 13 |
| Reasoning & Knowledge | 82 | 18 |
| Agentic & Tools | 94 | 8 |
| Specifications | ||
| Developer | Google DeepMind | Z.ai (Zhipu AI),Tsinghua University |
| Family | Gemini | — |
| Released | Aug 13, 2026 | Aug 5, 2025 |
| Parameters | — | 355B |
| Availability | API access | Open weights (unrestricted) |
| Context window | 1M | 131K |
| Price — $/M input | $0.75 | $0.60 |
| Price — $/M output | $3.75 | $2.20 |
| Inputs | text, image, video, file, audio | text |
| Outputs | text | text |
| Benchmarks | ||
| AIME 2024/2025 | 97% | 74% |
| APEX | 68% | — |
| ARC-AGI | 96% | — |
| ARC-AGI-2 | 85% | — |
| GPQA Diamond | 95% | 78% |
| Humanity's Last Exam | 39% | 8% |
| SciCode | 60% | 35% |
| SimpleQA Verified | 69% | — |
| WebDev Arena | 1587 | — |
| LiveCodeBench | — | 74% |
| MMLU-Pro | — | 84% |
| WeirdML | — | 41% |
| τ²-bench | — | 43% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Frequently asked questions
Which is cheaper, Gemini 3.7 Flash or GLM-4.5?
GLM-4.5 is cheaper on input tokens at $0.60 per million, versus $0.75 (representative OpenRouter pricing).
Which has a larger context window, Gemini 3.7 Flash or GLM-4.5?
Gemini 3.7 Flash supports up to 1M tokens, compared with 131K for the other.
Which is better for coding, Gemini 3.7 Flash or GLM-4.5?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Gemini 3.7 Flash ranks higher — 88th vs 19th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.