Model comparison
Qwen3.8 27B vs Qwen3-Coder 480B-A35B Instruct
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 37.7 on the Noometry Index.
Last verified . 18 shared benchmarks.
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Qwen3.8 27B scores higher in 8 categories and Qwen3-Coder 480B-A35B Instruct in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen3.8 27B leads 50.3 to 36.2.
- Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1.50 / $7.50 for Qwen3-Coder 480B-A35B Instruct.
Side by side
| Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 45.3 | 37.7 |
| Released | 2026-08-14 | 2025-04 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 33K | 66K |
| Input $ / M tokens | $0.99 | $1.50 |
| Output $ / M tokens | $1.49 | $7.50 |
| Results tracked | 28 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Qwen3.8 27B: 50.3 (#42), Qwen3-Coder 480B-A35B Instruct: 36.2 (#199)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena WebDev | 1592 | 1275 |
| LMArena Coding | 1487 | 1413 |
| SWE-bench Verified (bash only) | — | 55.4% |
| SciCode | 46.6% | — |
| GSO | — | 4.9% |
| WeirdML | — | 41.2% |
| ALE-Bench | — | 461.45 |
| AlgoTune | — | 1.44 |
Agentic & Tool Use Not comparable
Qwen3.8 27B: —, Qwen3-Coder 480B-A35B Instruct: 23.7 (#116)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |
Reasoning Qwen3.8 27B leads
Qwen3.8 27B: 31.5 (#71), Qwen3-Coder 480B-A35B Instruct: 24.7 (#141)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1459 | 1372 |
| Kagi LLM Benchmark | — | 49.5% |
| NYT Connections (extended) | 54.5% | — |
| CritPt | 5.4% | — |
| DTBench | 88% | — |
| LMCA | 41.4% | — |
| Surface Evolver Bench | 45% | — |
| Epoch Capabilities Index | 149.38 | — |
Math Too close to call
Qwen3.8 27B: 35.7 (#139), Qwen3-Coder 480B-A35B Instruct: 34.9 (#151)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1453 | 1364 |
| ProofBench | 16% | — |
Knowledge Qwen3.8 27B leads
Qwen3.8 27B: 40.9 (#105), Qwen3-Coder 480B-A35B Instruct: 36.4 (#159)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1482 | 1337 |
Multimodal Not comparable
Qwen3.8 27B: 40.5 (#36), Qwen3-Coder 480B-A35B Instruct: —
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Vision | 1270 | — |
Multilingual Qwen3.8 27B leads
Qwen3.8 27B: 54.0 (#54), Qwen3-Coder 480B-A35B Instruct: 47.8 (#143)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1431 | 1345 |
| LMArena Chinese | 1504 | 1356 |
| LMArena French | 1468 | 1394 |
| LMArena German | 1435 | 1324 |
| LMArena Japanese | 1392 | 1309 |
| LMArena Korean | 1389 | 1307 |
| LMArena Russian | 1421 | 1367 |
| LMArena Spanish | 1444 | 1360 |
Instruction Following Qwen3.8 27B leads
Qwen3.8 27B: 75.8 (#54), Qwen3-Coder 480B-A35B Instruct: 71.8 (#142)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1436 | 1356 |
Long Context Qwen3.8 27B leads
Qwen3.8 27B: 44.3 (#66), Qwen3-Coder 480B-A35B Instruct: 42.0 (#126)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1448 | 1379 |
Writing & Preference Qwen3.8 27B leads
Qwen3.8 27B: 65.9 (#41), Qwen3-Coder 480B-A35B Instruct: 55.5 (#142)
| Benchmark | Qwen3.8 27B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1441 | 1356 |
| LMArena Creative Writing | 1382 | 1333 |
| LMArena Multi-Turn | 1443 | 1365 |
| EQ-Bench Creative Writing | 1671 | — |
Frequently asked questions
Is Qwen3.8 27B better than Qwen3-Coder 480B-A35B Instruct?
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 37.7 on the Noometry Index.
Which is cheaper, Qwen3.8 27B or Qwen3-Coder 480B-A35B Instruct?
Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.
Is Qwen3.8 27B or Qwen3-Coder 480B-A35B Instruct better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.3 versus 36.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Qwen3.8 27B and Qwen3-Coder 480B-A35B Instruct share?
18 benchmarks have published results for both models. Qwen3.8 27B has 28 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.