Compare / Qwen3-235B-A22B-Instruct-2507 vs GLM-4.5-Air
Qwen3-235B-A22B-Instruct-2507 vs GLM-4.5-Air
Which is cheaper, which is more capable, and which does better at each kind of work, from 17 results on 3 sources that measured both models.
Of the 17 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 9 October 2026 (catalogue release 2026-10-09-dd620e778ac5). See also Alibaba vs Z.ai, every model from both labs.
Qwen3-235B-A22B-Instruct-2507 or GLM-4.5-Air? The short answer
They score and cost about the same: pick on the results below that match your work.
- They cost about the same: $0.30 and $0.32 per million tokens (three input to one output).
- They score about the same on the Artificial Analysis Intelligence Index (12.0 and 11.1).
- Qwen3-235B-A22B-Instruct-2507 is newer: released 21 Jul 2025, against 20 Jul 2025 for GLM-4.5-Air.
- On third-party benchmarks, on the results both models have: Qwen3-235B-A22B-Instruct-2507 does better for reasoning and knowledge, writing and sticking to the facts.
Head to head
Every fact side by side
| Qwen3-235B-A22B-Instruct-2507 | GLM-4.5-Air | |
|---|---|---|
| Price per million tokens | $0.15 in · $0.75 out | $0.14 in · $0.86 out |
| Blended (3 in : 1 out) | $0.30 | $0.32 |
| Cheapest host, blended | $0.15 (GMICloud, FP8) | $0.31 (Novita, BF16) |
| Intelligence Index | 12.0 | 11.1 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 262,144 tokens | 131,072 tokens |
| Released | 21 Jul 2025 (newer) | 20 Jul 2025 |
| Weights | Open | Open |
| Developer based in | China | China |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Qwen3-235B-A22B-Instruct-2507 | GLM-4.5-Air |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 109 · 169 of 387 measured | 1,422 | 1,373 |
| Overall · Text factualityArena rating, higher is better 94 · 137 of 163 measured | 1,431 | 1,395 |
| Business, management and finance · TextArena rating, higher is better 97 · 164 of 380 measured | 1,432 | 1,372 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 153 · 163 of 243 measured | 12.0 | 11.1 |
| GPQA Diamond% of questions, higher is better 136 · 145 of 231 measured | 75.3% | 73.3% |
| Humanity's Last Exam% of questions, higher is better 144 · 170 of 242 measured | 11.1% | 7.0% |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 80 · 83 of 230 measured | 49.0 | 47.0no reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 69 · 43 of 230 measured | 15.0% | 10.0%no reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 64 · 155 of 230 measured | 0.36 | 0.33 |
Show all 17 results (8 more)
| Measure | Qwen3-235B-A22B-Instruct-2507 | GLM-4.5-Air |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results | ||
| Creative writing · TextArena rating, higher is better 122 · 179 of 385 measured | 1,378 | 1,327 |
| Expert prompts · TextArena rating, higher is better 103 · 167 of 338 measured | 1,447 | 1,389 |
| Instruction following · TextArena rating, higher is better 104 · 169 of 387 measured | 1,414 | 1,361 |
| Writing, literature and language · TextArena rating, higher is better 115 · 180 of 386 measured | 1,396 | 1,341 |
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| IFBench% of instructions, higher is better 113 · 149 of 195 measured | 46.1% | 37.6% |
| Long-context reasoning (AA-LCR)% of questions, higher is better 177 · 160 of 233 measured | 33.9% | 46.7% |
| Terminal-Bench Hard% of tasks, higher is better 110 · 91 of 190 measured | 15.2% | 20.5% |
| τ²-bench telecom% of tasks, higher is better 127 · 107 of 193 measured | 33.3% | 46.5% |
Compare them with other models
Quick answers
Is Qwen3-235B-A22B-Instruct-2507 better than GLM-4.5-Air?
They score and cost about the same: pick on the results below that match your work.
Which is cheaper, Qwen3-235B-A22B-Instruct-2507 or GLM-4.5-Air?
They cost about the same: $0.30 and $0.32 per million tokens (three input to one output).
Which is better for reasoning and knowledge, Qwen3-235B-A22B-Instruct-2507 or GLM-4.5-Air?
Qwen3-235B-A22B-Instruct-2507 ranks 129 of 198 models on sale for hard questions across science, maths and general knowledge, against 139 for GLM-4.5-Air.
Which is better for writing, Qwen3-235B-A22B-Instruct-2507 or GLM-4.5-Air?
Qwen3-235B-A22B-Instruct-2507 ranks 85 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 105 for GLM-4.5-Air.
Which is better for sticking to the facts, Qwen3-235B-A22B-Instruct-2507 or GLM-4.5-Air?
Qwen3-235B-A22B-Instruct-2507 ranks 98 of 152 models on sale for summarising without inventing things, and factual answers, against 133 for GLM-4.5-Air.
Which has the bigger context window, Qwen3-235B-A22B-Instruct-2507 or GLM-4.5-Air?
Qwen3-235B-A22B-Instruct-2507: 262,144 tokens, against 131,072 for GLM-4.5-Air.
Run Qwen3-235B-A22B-Instruct-2507 and GLM-4.5-Air on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.