Model comparison
Qwen3.8 27B vs Wizardlm 70b
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 33.0 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)
MicrosoftSummary
- They share 12 benchmarks with published results for both. Qwen3.8 27B scores higher in 7 categories and Wizardlm 70b in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.8 27B leads 65.8 to 34.8.
Side by side
| Qwen3.8 27B | Wizardlm 70b | |
|---|---|---|
| Provider | Alibaba (Qwen) | Microsoft |
| Noometry Index | 46.0 | 33.0 |
| Released | 2026-08-14 | — |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.99 | — |
| Output $ / M tokens | $1.49 | — |
| Results tracked | 31 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Qwen3.8 27B: 50.5 (#44), Wizardlm 70b: 31.4 (#285)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Coding | 1482 | 1081 |
| LMArena WebDev | 1593 | — |
| SciCode | 46.6% | — |
Agentic & Tool Use Not comparable
Qwen3.8 27B: 32.9 (#57), Wizardlm 70b: —
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| APEX-Agents | 47.5% | — |
Reasoning Qwen3.8 27B leads
Qwen3.8 27B: 41.0 (#54), Wizardlm 70b: 20.7 (#234)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Hard Prompts | 1460 | 1079 |
| ARC-AGI-2 | 42.4% | — |
| NYT Connections (extended) | 54.5% | — |
| ARC-AGI-1 | 87.5% | — |
| CritPt | 5.4% | — |
| DTBench | 88% | — |
| LMCA | 41.4% | — |
| Surface Evolver Bench | 45% | — |
| Epoch Capabilities Index | 149.38 | — |
Math Qwen3.8 27B leads
Qwen3.8 27B: 37.1 (#161), Wizardlm 70b: 32.2 (#218)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Math | 1456 | 1116 |
| ProofBench | 16% | — |
Knowledge Not comparable
Qwen3.8 27B: 41.6 (#109), Wizardlm 70b: —
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Expert | 1482 | — |
Multimodal Not comparable
Qwen3.8 27B: 41.3 (#37), Wizardlm 70b: —
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Vision | 1271 | — |
Multilingual Qwen3.8 27B leads
Qwen3.8 27B: 53.7 (#60), Wizardlm 70b: 29.6 (#265)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Non-English | 1430 | 1078 |
| LMArena Chinese | 1504 | 1052 |
| LMArena German | 1438 | 1083 |
| LMArena Russian | 1415 | 1155 |
| LMArena French | 1465 | — |
| LMArena Japanese | 1384 | — |
| LMArena Korean | 1393 | — |
| LMArena Spanish | 1448 | — |
Instruction Following Qwen3.8 27B leads
Qwen3.8 27B: 75.8 (#53), Wizardlm 70b: 56.3 (#273)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Instruction Following | 1439 | 1093 |
Long Context Qwen3.8 27B leads
Qwen3.8 27B: 44.3 (#70), Wizardlm 70b: 33.3 (#263)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Longer Query | 1450 | 1097 |
Writing & Preference Qwen3.8 27B leads
Qwen3.8 27B: 65.8 (#43), Wizardlm 70b: 34.8 (#269)
| Benchmark | Qwen3.8 27B | Wizardlm 70b |
|---|---|---|
| LMArena Text | 1441 | 1120 |
| LMArena Creative Writing | 1384 | 1149 |
| LMArena Multi-Turn | 1441 | 1108 |
| EQ-Bench Creative Writing | 1671 | — |
Frequently asked questions
Is Qwen3.8 27B better than Wizardlm 70b?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 33.0 on the Noometry Index.
Is Qwen3.8 27B or Wizardlm 70b better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 31.4 in the Noometry coding category.
How many benchmarks do Qwen3.8 27B and Wizardlm 70b share?
12 benchmarks have published results for both models. Qwen3.8 27B has 31 scored results on Noometry and Wizardlm 70b has 12.