Model comparison
Llama 3.2 1B vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 20.1 on the Noometry Index. Llama 3.2 1B costs 17× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Llama 3.2 1B scores higher in 1 category and Qwen3 235B-A22B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.6 to 7.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.6% for Llama 3.2 1B and 86.7% for Qwen3 235B-A22B.
- Llama 3.2 1B is cheaper at $0.027 / $0.20 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- Qwen3 235B-A22B accepts more context: 131K tokens versus 60K.
Side by side
| Llama 3.2 1B | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 20.1 | 43.5 |
| Released | 2024-09-24 | 2025-04 |
| Weights | Open | Open |
| Context window | 60K | 131K |
| Max output | 54K | 16K |
| Input $ / M tokens | $0.027 | $0.70 |
| Output $ / M tokens | $0.20 | $2.80 |
| Results tracked | 22 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Llama 3.2 1B: 21.1 (#338), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Coding | 1070 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
| BigCodeBench Instruct | 8.2% | — |
| BigCodeBench Complete | 11.3% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Llama 3.2 1B: 14.6 (#150), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 10.8% | 52.1% |
| BALROG | 6.6% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Too close to call
Llama 3.2 1B: 16.2 (#308), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| Chess Puzzles | 0% | 12% |
| LMArena Hard Prompts | 1044 | 1433 |
| Epoch Capabilities Index | 101.99 | 143.85 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 29.3% |
| ForecastBench | — | 59.7 |
Math Qwen3 235B-A22B leads
Llama 3.2 1B: 10.4 (#313), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 86.7% |
| LMArena Math | 1086 | 1432 |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Llama 3.2 1B: 7.2 (#312), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 23.9% | 80.1% |
| LMArena Expert | 1007 | 1463 |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multilingual Qwen3 235B-A22B leads
Llama 3.2 1B: 23.8 (#292), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 973 | 1409 |
| LMArena Chinese | 959 | 1481 |
| LMArena German | 1014 | 1433 |
| LMArena Russian | 941 | 1411 |
| LMArena French | — | 1445 |
| LMArena Japanese | — | 1399 |
| LMArena Korean | — | 1391 |
| LMArena Spanish | — | 1430 |
Instruction Following Qwen3 235B-A22B leads
Llama 3.2 1B: 52.4 (#290), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1031 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Llama 3.2 1B: 31.9 (#274), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1050 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Llama 3.2 1B: 21.3 (#310), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Llama 3.2 1B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1055 | 1419 |
| LMArena Creative Writing | 1033 | 1384 |
| EQ-Bench Creative Writing | 200 | 1366 |
| LMArena Multi-Turn | 1030 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |
Frequently asked questions
Is Llama 3.2 1B better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 20.1 on the Noometry Index. Llama 3.2 1B costs 17× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.
Which is cheaper, Llama 3.2 1B or Qwen3 235B-A22B?
Llama 3.2 1B is cheaper. It lists at $0.027 per million input tokens and $0.20 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is Llama 3.2 1B or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 21.1 in the Noometry coding category.
Which has the bigger context window?
Qwen3 235B-A22B does, with 131K tokens against 60K.
How many benchmarks do Llama 3.2 1B and Qwen3 235B-A22B share?
19 benchmarks have published results for both models. Llama 3.2 1B has 22 scored results on Noometry and Qwen3 235B-A22B has 49.