Compare / Qwen3.5-27B vs GPT-5.6 Luna
Qwen3.5-27B vs GPT-5.6 Luna
Which is cheaper, which is more capable, and which does better at each kind of work, from 15 results on 4 sources that measured both models.
Of the 15 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.
Qwen3.5-27B or GPT-5.6 Luna? The short answer
Pick GPT-5.6 Luna: it scores higher (Intelligence Index 37.3 against 22.9) and costs less.
- GPT-5.6 Luna is cheaper: $0.45 against $0.74 per million tokens, 39% less.
- GPT-5.6 Luna scores higher on the Artificial Analysis Intelligence Index: 37.3 against 22.9.
- GPT-5.6 Luna is newer: released 9 Jul 2026, against 24 Feb 2026 for Qwen3.5-27B.
- On third-party benchmarks, on the results both models have: GPT-5.6 Luna does better for agents and tool use, reasoning and knowledge, writing and sticking to the facts.
Head to head
Every fact side by side
| Qwen3.5-27B | GPT-5.6 Luna | |
|---|---|---|
| Price per million tokens | $0.27 in · $2.16 out | $0.20 in · $1.20 out |
| Blended (3 in : 1 out) | $0.74 | $0.45 |
| Cheapest host, blended | $0.54 (Alibaba) | $0.45 (Azure) |
| Cached input, per million | – | $0.020 cached input |
| Intelligence Index | 22.9 | 37.3 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 262,144 tokens | 1,050,000 tokens |
| Released | 24 Feb 2026 | 9 Jul 2026 (newer) |
| Weights | Open | Closed |
| 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.
From our research
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 | Qwen3.5-27B | GPT-5.6 Luna |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 129 · 64 of 387 measured | 1,409 | 1,453xhigh reasoning |
| Overall · Text factualityArena rating, higher is better 112 · 40 of 163 measured | 1,420 | 1,460xhigh reasoning |
| Business, management and finance · TextArena rating, higher is better 128 · 47 of 380 measured | 1,408 | 1,463xhigh reasoning |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 86 · 37 of 243 measured | 22.9 | 37.3max reasoning |
| GPQA Diamond% of questions, higher is better 74 · 31 of 231 measured | 85.8% | 91.1%max reasoning |
| Humanity's Last Exam% of questions, higher is better 95 · 43 of 242 measured | 23.9% | 39.5%max reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 99 · 67 of 230 measured | 42.4think-prefill reasoning | 53.7high reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 49 · 69 of 230 measured | 11.0%think-prefill reasoning | 15.0%xhigh reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 147 · 29 of 230 measured | 0.33no reasoning | 0.38xhigh reasoning |
| Vending-Bench 2 reported by Vending-Bench 2 · 1 result | ||
| Money after a yearUS dollars, higher is better 50 · 32 of 58 measured | $201.98 | $4,095 |
Show all 15 results (5 more)
| Measure | Qwen3.5-27B | GPT-5.6 Luna |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results | ||
| Creative writing · TextArena rating, higher is better 145 · 80 of 385 measured | 1,358 | 1,409xhigh reasoning |
| Expert prompts · TextArena rating, higher is better 112 · 51 of 338 measured | 1,441 | 1,491xhigh reasoning |
| Instruction following · TextArena rating, higher is better 122 · 63 of 387 measured | 1,401 | 1,445xhigh reasoning |
| Writing, literature and language · TextArena rating, higher is better 135 · 64 of 386 measured | 1,380 | 1,433xhigh reasoning |
| Artificial Analysis reported by Artificial Analysis · 1 result | ||
| Long-context reasoning (AA-LCR)% of questions, higher is better 63 · 10 of 233 measured | 77.7% | 83.7%max reasoning |
Compare them with other models
Quick answers
Is Qwen3.5-27B better than GPT-5.6 Luna?
Pick GPT-5.6 Luna: it scores higher (Intelligence Index 37.3 against 22.9) and costs less.
Which is cheaper, Qwen3.5-27B or GPT-5.6 Luna?
GPT-5.6 Luna is cheaper: $0.45 against $0.74 per million tokens, 39% less.
Which is better for agents and tool use, Qwen3.5-27B or GPT-5.6 Luna?
GPT-5.6 Luna ranks 50 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 145 for Qwen3.5-27B.
Which is better for reasoning and knowledge, Qwen3.5-27B or GPT-5.6 Luna?
GPT-5.6 Luna ranks 37 of 198 models on sale for hard questions across science, maths and general knowledge, against 82 for Qwen3.5-27B.
Which is better for writing, Qwen3.5-27B or GPT-5.6 Luna?
GPT-5.6 Luna ranks 59 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 98 for Qwen3.5-27B.
Which is better for sticking to the facts, Qwen3.5-27B or GPT-5.6 Luna?
GPT-5.6 Luna ranks 61 of 152 models on sale for summarising without inventing things, and factual answers, against 121 for Qwen3.5-27B.
Which has the bigger context window, Qwen3.5-27B or GPT-5.6 Luna?
GPT-5.6 Luna: 1,050,000 tokens, against 262,144 for Qwen3.5-27B.
Run Qwen3.5-27B and GPT-5.6 Luna on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.