Model comparison
GPT-6 Luna vs Qwen2.5-Coder-32B
GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.4 on the Noometry Index.
Last verified . 13 shared benchmarks.
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. GPT-6 Luna scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 33.3.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- GPT-6 Luna accepts more context: 1.05M tokens versus 33K.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Luna | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 53.3 | 33.4 |
| Released | 2026-09-22 | 2024-09-18 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 33K |
| Max output | 128K | 29K |
| Input $ / M tokens | $0.10 | $0.66 |
| Output $ / M tokens | $0.50 | $1 |
| Results tracked | 42 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 55.5 (#25), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1439 | 1276 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LMArena WebDev | 1581 | — |
| SciCode | 54.6% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 1,577 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
GPT-6 Luna: 33.3 (#54), Qwen2.5-Coder-32B: —
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| APEX-Agents | 44.3% | — |
| GDP.pdf | 23% | — |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1411 | 1251 |
| Epoch Capabilities Index | 156.28 | 119.49 |
| ARC-AGI-2 | 59.3% | — |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| CritPt | 19.4% | — |
| Chess Puzzles | 31% | — |
| LiveBench Reasoning | — | 42.1% |
| Mystery Game Puzzles | 7% | — |
| DTBench | 90.1% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 44.5% | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1416 | 1251 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 64% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge GPT-6 Luna leads
GPT-6 Luna: 57.0 (#41), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1444 | 1221 |
| GPQA Diamond | 90.5% | — |
| SimpleQA Verified | 41.4% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
GPT-6 Luna: 42.4 (#30), Qwen2.5-Coder-32B: —
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
Multilingual GPT-6 Luna leads
GPT-6 Luna: 50.5 (#117), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1386 | 1205 |
| LMArena Chinese | 1433 | 1222 |
| LMArena Russian | 1394 | 1228 |
| LMArena French | 1420 | — |
| LMArena German | 1369 | — |
| LMArena Japanese | 1369 | — |
| LMArena Korean | 1360 | — |
| LMArena Spanish | 1393 | — |
Instruction Following GPT-6 Luna leads
GPT-6 Luna: 74.3 (#99), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1409 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context GPT-6 Luna leads
GPT-6 Luna: 43.0 (#111), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1409 | 1251 |
Writing & Preference GPT-6 Luna leads
GPT-6 Luna: 58.3 (#119), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | GPT-6 Luna | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1391 | 1230 |
| LMArena Creative Writing | 1363 | 1174 |
| LMArena Multi-Turn | 1396 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is GPT-6 Luna better than Qwen2.5-Coder-32B?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.4 on the Noometry Index.
Which is cheaper, GPT-6 Luna or Qwen2.5-Coder-32B?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is GPT-6 Luna or Qwen2.5-Coder-32B better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 33K.
How many benchmarks do GPT-6 Luna and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Qwen2.5-Coder-32B has 31.