Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 4B/30B/235B | 9B/32B |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 256k | 128k |
| Benchmarks | ||
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 74.1% | 59.1% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | 87.8% | 84.5% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 92.3% | 63.6% |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 18.2% | — |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | — | 80.8% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 81.1% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 84.4% | — |
| Overview | ||
| Company | Qwen | Z.ai |
| Release date | Jul 21 2025 | Apr 14 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Qwen3-2507vsClaude Fable 5.2GLM-Z1-0414vsClaude Fable 5.2Qwen3-2507vsGPT-6.2 AstraGLM-Z1-0414vsGPT-6.2 AstraQwen3-2507vsGemini 4 NeonGLM-Z1-0414vsGemini 4 NeonQwen3-2507vsMuse EmberGLM-Z1-0414vsMuse EmberQwen3-2507vsGrok 4.7GLM-Z1-0414vsGrok 4.7Qwen3-2507vsDeepSeek-V4.1-Flash-1017GLM-Z1-0414vsDeepSeek-V4.1-Flash-1017Frequently asked questions
Qwen3-2507 leads GLM-Z1-0414 on 3 of the 3 benchmarks they both report (LiveCodeBench, IFEval, AIME 2025). GLM-Z1-0414 shipped 98 days before Qwen3-2507, so benchmark comparisons should account for the intervening progress.
Qwen3-2507 has 4B/30B/235B parameters, while GLM-Z1-0414 has 9B/32B. Context windows are 256k (Qwen3-2507) vs 128k (GLM-Z1-0414).
On LiveCodeBench, Qwen3-2507 leads at 74.1% vs GLM-Z1-0414 at 59.1%. On IFEval, Qwen3-2507 leads at 87.8% vs GLM-Z1-0414 at 84.5%. On AIME 2025, Qwen3-2507 leads at 92.3% vs GLM-Z1-0414 at 63.6%.