Model comparison
Mercury 2 vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 39.1 on the Noometry Index. Mercury 2 costs 8.0× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Mercury 2 scores higher in 0 categories and Qwen3.8 Max in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 54.4 to 23.8.
- The biggest single-benchmark swing is CritPt: 0.8% for Mercury 2 and 20% for Qwen3.8 Max.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 128K.
Side by side
| Mercury 2 | Qwen3.8 Max | |
|---|---|---|
| Provider | Inception | Alibaba (Qwen) |
| Noometry Index | 39.1 | 56.8 |
| Released | 2026-02-20 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 1M |
| Max output | 50K | 131K |
| Input $ / M tokens | $0.25 | $2 |
| Output $ / M tokens | $0.75 | $6 |
| Results tracked | 17 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Mercury 2: 33.5 (#255), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1171 | 1674 |
| SciCode | 38.7% | 53.2% |
| LMArena Coding | 1391 | 1502 |
| DeepSWE | — | 57.5% |
| FrontierSWE | — | 17.8% |
| WeirdML | 43.2% | — |
| ALE-Bench | 785.58 | — |
Agentic & Tool Use Not comparable
Mercury 2: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Mercury 2: 23.8 (#170), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| CritPt | 0.8% | 20% |
| LMArena Hard Prompts | 1362 | 1496 |
| NYT Connections (extended) | — | 88.3% |
| Chess Puzzles | — | 40% |
| Mystery Game Puzzles | — | 38% |
| DTBench | — | 92% |
| LMCA | — | 46.2% |
| Epoch Capabilities Index | — | 156.41 |
Math Not comparable
Mercury 2: —, Qwen3.8 Max: 73.2 (#20)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |
| LMArena Math | — | 1499 |
Knowledge Qwen3.8 Max leads
Mercury 2: 36.2 (#172), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1358 | 1507 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 47.3% |
| Vectara Hallucination Rate | 12.3% | — |
Multimodal Not comparable
Mercury 2: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Mercury 2: 46.6 (#157), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1331 | 1472 |
| LMArena Chinese | 1417 | 1538 |
| LMArena Russian | 1304 | 1481 |
| LMArena French | — | 1503 |
| LMArena German | — | 1483 |
| LMArena Japanese | — | 1467 |
| LMArena Korean | — | 1461 |
| LMArena Spanish | — | 1492 |
Instruction Following Qwen3.8 Max leads
Mercury 2: 70.2 (#165), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1329 | 1479 |
Long Context Qwen3.8 Max leads
Mercury 2: 40.5 (#154), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1330 | 1489 |
Writing & Preference Qwen3.8 Max leads
Mercury 2: 53.8 (#155), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Mercury 2 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1355 | 1483 |
| LMArena Creative Writing | 1289 | 1479 |
| LMArena Multi-Turn | 1358 | 1489 |
Frequently asked questions
Is Mercury 2 better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 39.1 on the Noometry Index. Mercury 2 costs 8.0× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, Mercury 2 or Qwen3.8 Max?
Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Mercury 2 or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 128K.
How many benchmarks do Mercury 2 and Qwen3.8 Max share?
14 benchmarks have published results for both models. Mercury 2 has 17 scored results on Noometry and Qwen3.8 Max has 39.