Compare / Qwen3.5-397B-A17B vs gpt-oss-120b
Qwen3.5-397B-A17B vs gpt-oss-120b
Which is cheaper, which is more capable, and which does better at each kind of work, from 25 results on 4 sources that measured both models.
Of the 25 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 OpenAI, every model from both labs.
Qwen3.5-397B-A17B or gpt-oss-120b? The short answer
Pick Qwen3.5-397B-A17B if quality matters most: it scores higher (Intelligence Index 21.4 against 11.6). Pick gpt-oss-120b if cost matters more: it costs 4.9 times less per token. gpt-oss-120b is also faster.
- gpt-oss-120b is cheaper: $0.26 against $1.29 per million tokens, 4.9 times less.
- Qwen3.5-397B-A17B scores higher on the Artificial Analysis Intelligence Index: 21.4 against 11.6.
- gpt-oss-120b writes faster: 188 tokens a second against 90.
- Qwen3.5-397B-A17B is newer: released 13 Feb 2026, against 5 Aug 2025 for gpt-oss-120b.
- On third-party benchmarks, on the results both models have: Qwen3.5-397B-A17B does better for agents and tool use, professional work, reasoning and knowledge, writing and sticking to the facts.
Head to head
Every fact side by side
| Qwen3.5-397B-A17B | gpt-oss-120b | |
|---|---|---|
| Price per million tokens | $0.55 in · $3.50 out | $0.15 in · $0.60 out |
| Blended (3 in : 1 out) | $1.29 | $0.26 |
| Cheapest host, blended | $0.88 (Alibaba) | $0.065 (CoreWeave, FP4) |
| Intelligence Index | 21.4 | 11.6 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | 90 | 188 |
| Context window | 262,144 tokens | 131,072 tokens |
| Released | 13 Feb 2026 (newer) | 5 Aug 2025 |
| Weights | Open | Open |
| Developer based in | China | the US |
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 | Qwen3.5-397B-A17B | gpt-oss-120b |
|---|---|---|
| APEX-Agents reported by APEX-Agents · 1 result | ||
| Tasks passed% of tasks, higher is better 36 · 37 of 37 measured | 24.9% | 4.4% |
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 80 · 187 of 387 measured | 1,442 | 1,352 |
| Overall · Text factualityArena rating, higher is better 70 · 142 of 163 measured | 1,445 | 1,385 |
| Business, management and finance · TextArena rating, higher is better 73 · 183 of 380 measured | 1,446 | 1,352 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 97 · 157 of 243 measured | 21.4no reasoning | 11.6high reasoning |
| GPQA Diamond% of questions, higher is better 53 · 118 of 231 measured | 89.3% | 78.2%high reasoning |
| Humanity's Last Exam% of questions, higher is better 77 · 110 of 242 measured | 29.0% | 19.6%high reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 77 · 116 of 230 measured | 50.1no reasoning | 38.5medium reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 134 · 224 of 230 measured | 23.0%no reasoning | 107.0%medium reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 164 · 217 of 230 measured | 0.33no reasoning | 0.29medium reasoning |
Show all 25 results (15 more)
| Measure | Qwen3.5-397B-A17B | gpt-oss-120b |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results | ||
| Creative writing · TextArena rating, higher is better 86 · 238 of 385 measured | 1,405 | 1,278 |
| Expert prompts · TextArena rating, higher is better 67 · 188 of 338 measured | 1,477 | 1,357 |
| Instruction following · TextArena rating, higher is better 78 · 204 of 387 measured | 1,434 | 1,323 |
| Writing, literature and language · TextArena rating, higher is better 80 · 215 of 386 measured | 1,421 | 1,309 |
| Artificial Analysis reported by Artificial Analysis · 11 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 61 · 87 of 130 measured | 51.3% | 26.2%high reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 66 · 66 of 112 measured | 0.0% | 0.0%high reasoning |
| τ-bench banking% of tasks, higher is better 64 · 67 of 121 measured | 13.4% | 12.8%high reasoning |
| Artificial Analysis Coding Indexindex score, higher is better 61 · 86 of 131 measured | 48.2 | 30.4high reasoning |
| IFBench% of instructions, higher is better 8 · 58 of 195 measured | 78.8% | 69.0%high reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 66 · 152 of 233 measured | 77.3% | 52.0%high reasoning |
| Output speedtokens per second, higher is better 42 · 12 of 74 measured | 90.0 | 188low reasoning |
| SciCode% of problems, higher is better 66 · 93 of 113 measured | 44.8% | 34.0%high reasoning |
| Terminal-Bench Hard% of tasks, higher is better 34 · 84 of 190 measured | 40.9% | 23.5%high reasoning |
| Time to first answer tokenseconds, lower is better 34 · 46 of 74 measured | 1.59no reasoning | 11.2low reasoning |
| τ²-bench telecom% of tasks, higher is better 16 · 86 of 193 measured | 95.6% | 65.8%high reasoning |
Compare them with other models
Quick answers
Is Qwen3.5-397B-A17B better than gpt-oss-120b?
Pick Qwen3.5-397B-A17B if quality matters most: it scores higher (Intelligence Index 21.4 against 11.6). Pick gpt-oss-120b if cost matters more: it costs 4.9 times less per token. gpt-oss-120b is also faster.
Which is cheaper, Qwen3.5-397B-A17B or gpt-oss-120b?
gpt-oss-120b is cheaper: $0.26 against $1.29 per million tokens, 4.9 times less.
Which is faster, Qwen3.5-397B-A17B or gpt-oss-120b?
gpt-oss-120b writes about 188 tokens a second on its usual API, against 90 for Qwen3.5-397B-A17B (Artificial Analysis).
Which is better for agents and tool use, Qwen3.5-397B-A17B or gpt-oss-120b?
Qwen3.5-397B-A17B ranks 70 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 120 for gpt-oss-120b.
Which is better for professional work, Qwen3.5-397B-A17B or gpt-oss-120b?
Qwen3.5-397B-A17B ranks 48 of 51 models on sale for real tasks from banking, consulting and law, business documents and freelance projects, against 49 for gpt-oss-120b.
Which is better for reasoning and knowledge, Qwen3.5-397B-A17B or gpt-oss-120b?
Qwen3.5-397B-A17B ranks 73 of 198 models on sale for hard questions across science, maths and general knowledge, against 114 for gpt-oss-120b.
Which is better for writing, Qwen3.5-397B-A17B or gpt-oss-120b?
Qwen3.5-397B-A17B ranks 69 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 133 for gpt-oss-120b.
Which has the bigger context window, Qwen3.5-397B-A17B or gpt-oss-120b?
Qwen3.5-397B-A17B: 262,144 tokens, against 131,072 for gpt-oss-120b.
Run Qwen3.5-397B-A17B and gpt-oss-120b on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.