Model comparison
Llama-3.3-70B-Instruct vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 7.9× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 32 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 0 categories and Qwen3 235B-A22B in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 235B-A22B leads 50.4 to 15.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 5.1% for Llama-3.3-70B-Instruct and 86.7% for Qwen3 235B-A22B.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- Qwen3 235B-A22B accepts more context: 131K tokens versus 128K.
Side by side
| Llama-3.3-70B-Instruct | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.6 | 43.5 |
| Released | 2024-12-06 | 2025-04 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 16K |
| Input $ / M tokens | $0.10 | $0.70 |
| Output $ / M tokens | $0.32 | $2.80 |
| Results tracked | 43 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 31.0 (#290), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 26% | 42.4% |
| WeirdML | 14.4% | 41% |
| LMArena Coding | 1268 | 1445 |
| Aider Polyglot | — | 59.6% |
| BigCodeBench Instruct | 46.9% | — |
| LiveBench Coding | 36.6% | — |
| BigCodeBench Complete | 57.5% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 25.8 (#105), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | 52.1% |
| BALROG | 23% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 14.1 (#327), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| SimpleBench | 19.9% | 31% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1257 | 1433 |
| DTBench | 59.5% | 80.3% |
| LMCA | 17.5% | 29.3% |
| Epoch Capabilities Index | 127.33 | 143.85 |
| ForecastBench | 58.6 | 59.7 |
| ARC-AGI-2 | — | 1.3% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| Chess Puzzles | — | 12% |
| LiveBench Reasoning | 50.8% | — |
| Mystery Game Puzzles | — | 9% |
| LiveBench Data Analysis | 49.5% | — |
| LiveBench | 50.2% | — |
Math Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 15.3 (#298), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.1% | 86.7% |
| LMArena Math | 1267 | 1432 |
| MATH Level 5 | 41.6% | 68.9% |
| Omni-MATH | — | 71.8% |
| LiveBench Math | 42.2% | — |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 30.6 (#226), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 47.4% | 80.1% |
| Confabulations | 22.8% | 15.6% |
| Vectara Hallucination Rate | 4.1% | 9.3% |
| LMArena Expert | 1225 | 1463 |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| GPQA (HELM) | — | 72.7% |
| MMLU | 86.3% | — |
Multilingual Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 39.9 (#220), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1236 | 1409 |
| LMArena Chinese | 1217 | 1481 |
| LMArena French | 1281 | 1445 |
| LMArena German | 1251 | 1433 |
| LMArena Japanese | 1150 | 1399 |
| LMArena Korean | 1143 | 1391 |
| LMArena Russian | 1252 | 1411 |
| LMArena Spanish | 1270 | 1430 |
Instruction Following Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 71.1 (#157), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1242 | 1408 |
| LiveBench Instruction Following | 82.7% | — |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 26.4 (#295), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| Fiction.LiveBench | 33.3% | 75% |
| LMArena Longer Query | 1256 | 1426 |
Writing & Preference Qwen3 235B-A22B leads
Llama-3.3-70B-Instruct: 47.6 (#207), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1274 | 1419 |
| LMArena Creative Writing | 1250 | 1384 |
| LMArena Multi-Turn | 1280 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 7.9× 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-3.3-70B-Instruct or Qwen3 235B-A22B?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is Llama-3.3-70B-Instruct or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Qwen3 235B-A22B does, with 131K tokens against 128K.
How many benchmarks do Llama-3.3-70B-Instruct and Qwen3 235B-A22B share?
32 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Qwen3 235B-A22B has 49.