Model comparison
Olmo 3 32b Think vs Qwen3.5-Flash
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 38.7 on the Noometry Index.
Last verified . 14 shared benchmarks.
Allen Institute for AI (Ai2)
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Olmo 3 32b Think scores higher in 1 category and Qwen3.5-Flash in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Qwen3.5-Flash leads 50.5 to 41.2.
- Olmo 3 32b Think has downloadable open weights; the other is API-only.
Side by side
| Olmo 3 32b Think | Qwen3.5-Flash | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | Alibaba (Qwen) |
| Noometry Index | 38.7 | 42.5 |
| Released | — | 2026-02-23 |
| Weights | Open | Proprietary |
| Context window | — | 1M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 14 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Olmo 3 32b Think leads
Olmo 3 32b Think: 38.6 (#172), Qwen3.5-Flash: 34.2 (#242)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Coding | 1319 | 1412 |
| LMArena WebDev | — | 1244 |
| ALE-Bench | — | 221.8 |
Agentic & Tool Use Not comparable
Olmo 3 32b Think: —, Qwen3.5-Flash: —
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| Vending-Bench 2 | — | 462.69 |
Reasoning Qwen3.5-Flash leads
Olmo 3 32b Think: 25.9 (#140), Qwen3.5-Flash: 33.7 (#72)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Hard Prompts | 1302 | 1403 |
| Chess Puzzles | — | 21% |
| Mystery Game Puzzles | — | 20% |
| DTBench | — | 82.9% |
| LMCA | — | 29.1% |
| Epoch Capabilities Index | — | 143.98 |
Math Too close to call
Olmo 3 32b Think: 36.5 (#165), Qwen3.5-Flash: 37.4 (#158)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Math | 1316 | 1407 |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| OTIS Mock AIME 2024-2025 | — | 84.4% |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3.5-Flash leads
Olmo 3 32b Think: 35.0 (#190), Qwen3.5-Flash: 43.2 (#93)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Expert | 1273 | 1407 |
| GPQA Diamond | — | 82.3% |
| SimpleQA Verified | — | 20.3% |
| Vectara Hallucination Rate | — | 10.5% |
Multilingual Qwen3.5-Flash leads
Olmo 3 32b Think: 41.2 (#210), Qwen3.5-Flash: 50.5 (#121)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1255 | 1385 |
| LMArena Chinese | 1300 | 1446 |
| LMArena French | 1291 | 1412 |
| LMArena German | 1290 | 1390 |
| LMArena Russian | 1254 | 1379 |
| LMArena Japanese | — | 1368 |
| LMArena Korean | — | 1344 |
| LMArena Spanish | — | 1400 |
Instruction Following Qwen3.5-Flash leads
Olmo 3 32b Think: 67.2 (#198), Qwen3.5-Flash: 72.6 (#139)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1275 | 1374 |
Long Context Qwen3.5-Flash leads
Olmo 3 32b Think: 39.4 (#182), Qwen3.5-Flash: 42.4 (#124)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1296 | 1392 |
Writing & Preference Qwen3.5-Flash leads
Olmo 3 32b Think: 49.1 (#193), Qwen3.5-Flash: 57.9 (#122)
| Benchmark | Olmo 3 32b Think | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1300 | 1397 |
| LMArena Creative Writing | 1256 | 1343 |
| LMArena Multi-Turn | 1290 | 1393 |
Frequently asked questions
Is Olmo 3 32b Think better than Qwen3.5-Flash?
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 38.7 on the Noometry Index.
Is Olmo 3 32b Think or Qwen3.5-Flash better for coding?
Olmo 3 32b Think scores higher on coding benchmarks: 38.6 versus 34.2 in the Noometry coding category.
How many benchmarks do Olmo 3 32b Think and Qwen3.5-Flash share?
14 benchmarks have published results for both models. Olmo 3 32b Think has 14 scored results on Noometry and Qwen3.5-Flash has 32.