Model comparison
GLM-4.5-Air vs DeepSeek-R1-Distill-Qwen-32B
- Input
- $0.20
- Output
- $1.10
- Context
- 128K
- Released
- 2025-07-28
- Input
- $0.69 per 1M tokens
- Output
- $0.69 per 1M tokens
- Context
- 131,072 tokens (128K)
- Released
- 2025-01-20
GLM-4.5-Air scores higher on 4 of the 4 benchmarks both models report; GLM-4.5-Air is cheaper on input tokens ($0.2 vs $0.7 per 1M, 3.5×); DeepSeek-R1-Distill-Qwen-32B has the larger context window (131K vs 128K).
Specs and pricing side by side
| Spec | GLM-4.5-Air | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| Maker | Z.ai (Zhipu / GLM) | DeepSeek |
| Released | 2025-07-28 | 2025-01-20 |
| Context window | 128K | 131,072 tokens (128K) |
| Max output | 96K | 32,768 tokens (recommended max generation… |
| Input price | $0.20 | $0.69 per 1M tokens |
| Output price | $1.10 | $0.69 per 1M tokens |
| Parameters | 106B total · 12B active | ~32.8B (dense) |
| License | MIT | MIT License (model weights). The base Qwen2.5-32B it was distilled from is originally licensed under Apache 2.0. |
| Weights | Open weights | Open weightssame |
| Modalities | Text | text |
| Knowledge cutoff | Not disclosed | Inherits the Qwen2.5 base pretraining cutoff… |
Benchmarks both models report
| Benchmark | GLM-4.5-Air | DeepSeek-R1-Distill-Qwen-32B |
|---|---|---|
| AIME 2024 | 89.4% | 72.6% |
| MATH-500 | 98.1% | 94.3% |
| GPQA Diamond | 71.72% | 62.1% |
| LiveCodeBench | 70.7% | 57.2% |
Scores are as published by each maker or its cited source; harness, reasoning effort and date can differ between rows, so treat small gaps as ties. Sources are listed on each model page.
Which one should you pick?
Pick GLM-4.5-Air if…
- You care about AIME 2024, MATH-500, GPQA Diamond (it scores higher on 4 shared benchmarks).
- Cost matters: input tokens cost 3.5× less ($0.2 vs $0.7 per 1M).
- You need Text input, which DeepSeek-R1-Distill-Qwen-32B lacks.
- Your use case is: agentic coding assistants and autonomous coding agents that call tools over long horizons.
Pick DeepSeek-R1-Distill-Qwen-32B if…
- You need the longer context (131K vs 128K).
- You need text input, which GLM-4.5-Air lacks.
- Your use case is: local and on-prem reasoning assistants where data cannot leave the building.
FAQ
Which is better, GLM-4.5-Air or DeepSeek-R1-Distill-Qwen-32B?
GLM-4.5-Air scores higher on 4 of the 4 benchmarks both models report; GLM-4.5-Air is cheaper on input tokens ($0.2 vs $0.7 per 1M, 3.5×); DeepSeek-R1-Distill-Qwen-32B has the larger context window (131K vs 128K). The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is GLM-4.5-Air?
GLM-4.5-Air: $0.20 input, $1.10 output. DeepSeek-R1-Distill-Qwen-32B: $0.69 per 1M tokens input, $0.69 per 1M tokens output (list API prices as published by each maker).
What are the context windows of GLM-4.5-Air and DeepSeek-R1-Distill-Qwen-32B?
GLM-4.5-Air offers 128K with up to 96K output; DeepSeek-R1-Distill-Qwen-32B offers 131,072 tokens (128K) with up to 32,768 tokens (recommended max generation length; some hosted providers cap lower) output.
Is GLM-4.5-Air newer than DeepSeek-R1-Distill-Qwen-32B?
GLM-4.5-Air was released 2025-07-28; DeepSeek-R1-Distill-Qwen-32B was released 2025-01-20.
Go deeper
- GLM-4.5-Air: full specs, pricing and benchmarks
- DeepSeek-R1-Distill-Qwen-32B: full specs, pricing and benchmarks
- glm 4 5 air vs minimax m1
- gpt oss 120b vs glm 4 5 air
- deepseek v3 2 vs glm 4 5 air
- grok 3 vs deepseek r1 distill qwen 32b
- glm 4 5 air vs deepseek r1
- glm 4 5 air vs deepseek r1 distill qwen 14b
- All model comparisons