Compare / Qwen3-32B vs GLM-4.6
Qwen3-32B vs GLM-4.6
Which is cheaper, which is more capable, and which does better at each kind of work, from 29 results on 5 sources that measured both models.
Of the 29 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.
Qwen3-32B or GLM-4.6? The short answer
Pick GLM-4.6 if quality matters most: it scores higher (Intelligence Index 18.5 against 8.6). Pick Qwen3-32B if cost matters more: it costs 4.3 times less per token.
- Qwen3-32B is cheaper: $0.21 against $0.88 per million tokens, 4.3 times less.
- GLM-4.6 scores higher on the Artificial Analysis Intelligence Index: 18.5 against 8.6.
- GLM-4.6 is newer: released 30 Sep 2025, against 28 Apr 2025 for Qwen3-32B.
- On third-party benchmarks, on the results both models have: Qwen3-32B does better for sticking to the facts; GLM-4.6 for agents and tool use, reasoning and knowledge and writing.
Head to head
Every fact side by side
| Qwen3-32B | GLM-4.6 | |
|---|---|---|
| Price per million tokens | $0.14 in · $0.40 out | $0.50 in · $2.00 out |
| Blended (3 in : 1 out) | $0.21 | $0.88 |
| Cheapest host, blended | $0.13 (DeepInfra, FP8) | $0.76 (Venice, FP4) |
| Intelligence Index | 8.6 | 18.5 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 131,072 tokens | 204,800 tokens |
| Released | 28 Apr 2025 | 30 Sep 2025 (newer) |
| 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 | Qwen3-32B | GLM-4.6 |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 195 · 107 of 387 measured | 1,347 | 1,424 |
| Business, management and finance · TextArena rating, higher is better 203 · 120 of 380 measured | 1,340 | 1,412 |
| Creative writing · TextArena rating, higher is better 205 · 94 of 385 measured | 1,304 | 1,400 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 188 · 111 of 243 measured | 8.6thinking reasoning | 18.5thinking reasoning |
| GPQA Diamond% of questions, higher is better 167 · 120 of 231 measured | 66.8%thinking reasoning | 78.0%thinking reasoning |
| Humanity's Last Exam% of questions, higher is better 168 · 129 of 242 measured | 7.4%thinking reasoning | 14.5%thinking reasoning |
| Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 3 results | ||
| Overall accuracy · BFCL% correct, higher is better 24 · 4 of 81 measured | 48.7%bfcl fc | 72.4%thinking reasoning, bfcl fc |
| Irrelevance detection · BFCL% correct, higher is better 37 · 29 of 81 measured | 82.4%bfcl prompt | 85.0%thinking reasoning, bfcl fc |
| Memory · BFCL% correct, higher is better 21 · 5 of 81 measured | 26.7%bfcl fc | 55.7%thinking reasoning, bfcl fc |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 134 · 70 of 230 measured | 33.0no reasoning | 53.3thinking reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 99 · 57 of 230 measured | 19.0%no reasoning | 13.0%thinking reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 175 · 31 of 230 measured | 0.32no reasoning | 0.38thinking reasoning |
| Vectara Hallucination Leaderboard reported by Vectara Hallucination Leaderboard · 2 results | ||
| Hallucination rate · Vectara% of summaries, lower is better 24 · 53 of 104 measured | 5.9% | 9.5% |
| Answer rate · Vectara% of documents, higher is better 17 · 89 of 104 measured | 99.9% | 94.5% |
Show all 29 results (15 more)
| Measure | Qwen3-32B | GLM-4.6 |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Expert prompts · TextArena rating, higher is better 159 · 115 of 338 measured | 1,398 | 1,439 |
| Instruction following · TextArena rating, higher is better 198 · 108 of 387 measured | 1,331 | 1,412 |
| Writing, literature and language · TextArena rating, higher is better 217 · 102 of 386 measured | 1,308 | 1,405 |
| Artificial Analysis reported by Artificial Analysis · 7 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 111 · 63 of 130 measured | 5.2%thinking reasoning | 49.4%thinking reasoning |
| τ-bench banking% of tasks, higher is better 102 · 64 of 121 measured | 5.4%thinking reasoning | 13.4%thinking reasoning |
| Artificial Analysis Coding Indexindex score, higher is better 111 · 64 of 131 measured | 15.3thinking reasoning | 45.8thinking reasoning |
| IFBench% of instructions, higher is better 155 · 120 of 195 measured | 36.3%thinking reasoning | 43.4%thinking reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 222 · 145 of 233 measured | 0.0%no reasoning | 54.0%thinking reasoning |
| Terminal-Bench Hard% of tasks, higher is better 169 · 69 of 190 measured | 3.0%thinking reasoning | 28.8%no reasoning |
| τ²-bench telecom% of tasks, higher is better 135 · 74 of 193 measured | 29.8%thinking reasoning | 76.9%no reasoning |
| Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 5 results | ||
| Multi-turn tasks · BFCL% correct, higher is better 17 · 5 of 81 measured | 47.9%bfcl fc | 68.0%thinking reasoning, bfcl fc |
| Relevance detection · BFCL% correct, higher is better 8 · 49 of 81 measured | 93.8%bfcl fc | 75.0%thinking reasoning, bfcl fc |
| Single-turn calls (curated) · BFCL% correct, higher is better 3 · 22 of 81 measured | 90.3%bfcl prompt | 87.6%thinking reasoning, bfcl fc |
| Single-turn calls (user-contributed) · BFCL% correct, higher is better 3 · 6 of 81 measured | 82.0%bfcl fc | 80.9%thinking reasoning, bfcl fc |
| Web search · BFCL% correct, higher is better 28 · 8 of 81 measured | 26.0%bfcl prompt | 77.5%thinking reasoning, bfcl fc |
Compare them with other models
Quick answers
Is Qwen3-32B better than GLM-4.6?
Pick GLM-4.6 if quality matters most: it scores higher (Intelligence Index 18.5 against 8.6). Pick Qwen3-32B if cost matters more: it costs 4.3 times less per token.
Which is cheaper, Qwen3-32B or GLM-4.6?
Qwen3-32B is cheaper: $0.21 against $0.88 per million tokens, 4.3 times less.
Which is better for agents and tool use, Qwen3-32B or GLM-4.6?
GLM-4.6 ranks 55 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 113 for Qwen3-32B.
Which is better for reasoning and knowledge, Qwen3-32B or GLM-4.6?
GLM-4.6 ranks 108 of 198 models on sale for hard questions across science, maths and general knowledge, against 148 for Qwen3-32B.
Which is better for writing, Qwen3-32B or GLM-4.6?
GLM-4.6 ranks 79 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 139 for Qwen3-32B.
Which is better for sticking to the facts, Qwen3-32B or GLM-4.6?
Qwen3-32B ranks 22 of 152 models on sale for summarising without inventing things, and factual answers, against 70 for GLM-4.6.
Which has the bigger context window, Qwen3-32B or GLM-4.6?
GLM-4.6: 204,800 tokens, against 131,072 for Qwen3-32B.
Run Qwen3-32B and GLM-4.6 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.