Model comparison
Llama-3.3-70B-Instruct vs Qwen Plus
Qwen Plus is the stronger model overall, scoring 37.1 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 3.9× less per token, which makes it the better buy when Qwen Plus's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 2 categories and Qwen Plus in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen Plus leads 28.4 to 14.1.
- The biggest single-benchmark swing is MATH Level 5: 41.6% for Llama-3.3-70B-Instruct and 65.3% for Qwen Plus.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.40 / $1.20 for Qwen Plus.
- Qwen Plus accepts more context: 1M tokens versus 128K.
- Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.
Side by side
| Llama-3.3-70B-Instruct | Qwen Plus | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.6 | 37.1 |
| Released | 2024-12-06 | 2024-01-25 |
| Weights | Open | Proprietary |
| Context window | 128K | 1M |
| Max output | 4K | 33K |
| Input $ / M tokens | $0.10 | $0.40 |
| Output $ / M tokens | $0.32 | $1.20 |
| Results tracked | 43 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen Plus leads
Llama-3.3-70B-Instruct: 31.0 (#290), Qwen Plus: 38.9 (#167)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| LMArena Coding | 1268 | 1328 |
| SciCode | 26% | — |
| WeirdML | 14.4% | — |
| BigCodeBench Instruct | 46.9% | — |
| LiveBench Coding | 36.6% | — |
| BigCodeBench Complete | 57.5% | — |
Agentic & Tool Use Not comparable
Llama-3.3-70B-Instruct: 25.8 (#105), Qwen Plus: —
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |
Reasoning Qwen Plus leads
Llama-3.3-70B-Instruct: 14.1 (#327), Qwen Plus: 28.4 (#107)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1317 |
| DTBench | 59.5% | 81.1% |
| LMCA | 17.5% | 24% |
| SimpleBench | 19.9% | — |
| Kagi LLM Benchmark | — | 63.3% |
| CritPt | 0% | — |
| LiveBench Reasoning | 50.8% | — |
| LiveBench Data Analysis | 49.5% | — |
| Epoch Capabilities Index | 127.33 | — |
| ForecastBench | 58.6 | — |
| LiveBench | 50.2% | — |
Math Qwen Plus leads
Llama-3.3-70B-Instruct: 15.3 (#298), Qwen Plus: 23.3 (#271)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.1% | 17.8% |
| LMArena Math | 1267 | 1326 |
| MATH Level 5 | 41.6% | 65.3% |
| LiveBench Math | 42.2% | — |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 30.6 (#226), Qwen Plus: 27.4 (#251)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| GPQA Diamond | 47.4% | 48.1% |
| LMArena Expert | 1225 | 1328 |
| Confabulations | 22.8% | — |
| Vectara Hallucination Rate | 4.1% | — |
| MMLU | 86.3% | — |
Multilingual Qwen Plus leads
Llama-3.3-70B-Instruct: 39.9 (#220), Qwen Plus: 45.1 (#175)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| LMArena Non-English | 1236 | 1310 |
| LMArena Chinese | 1217 | 1347 |
| LMArena Japanese | 1150 | 1251 |
| LMArena Russian | 1252 | 1323 |
| LMArena French | 1281 | — |
| LMArena German | 1251 | — |
| LMArena Korean | 1143 | — |
| LMArena Spanish | 1270 | — |
Instruction Following Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 71.1 (#157), Qwen Plus: 68.8 (#181)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| LMArena Instruction Following | 1242 | 1303 |
| LiveBench Instruction Following | 82.7% | — |
Long Context Qwen Plus leads
Llama-3.3-70B-Instruct: 26.4 (#295), Qwen Plus: 40.3 (#158)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| LMArena Longer Query | 1256 | 1324 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Qwen Plus leads
Llama-3.3-70B-Instruct: 47.6 (#207), Qwen Plus: 52.2 (#176)
| Benchmark | Llama-3.3-70B-Instruct | Qwen Plus |
|---|---|---|
| LMArena Text | 1274 | 1326 |
| LMArena Creative Writing | 1250 | 1293 |
| LMArena Multi-Turn | 1280 | 1336 |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Qwen Plus?
Qwen Plus is the stronger model overall, scoring 37.1 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 3.9× less per token, which makes it the better buy when Qwen Plus's lead doesn't matter for your workload.
Which is cheaper, Llama-3.3-70B-Instruct or Qwen Plus?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Qwen Plus lists at $0.40 and $1.20.
Is Llama-3.3-70B-Instruct or Qwen Plus better for coding?
Qwen Plus scores higher on coding benchmarks: 38.9 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Qwen Plus does, with 1M tokens against 128K.
How many benchmarks do Llama-3.3-70B-Instruct and Qwen Plus share?
18 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Qwen Plus has 20.