Model comparison
Llama-3.3-70B-Instruct vs Qwen3.5 35B-A3B
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 4.4× less per token, which makes it the better buy when Qwen3.5 35B-A3B's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 0 categories and Qwen3.5 35B-A3B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.5 35B-A3B leads 39.9 to 15.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 5.1% for Llama-3.3-70B-Instruct and 70% for Qwen3.5 35B-A3B.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.25 / $2 for Qwen3.5 35B-A3B.
- Qwen3.5 35B-A3B accepts more context: 262K tokens versus 128K.
Side by side
| Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.6 | 42.0 |
| Released | 2024-12-06 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 66K |
| Input $ / M tokens | $0.10 | $0.25 |
| Output $ / M tokens | $0.32 | $2 |
| Results tracked | 43 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 31.0 (#290), Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| SciCode | 26% | 29.3% |
| LMArena Coding | 1268 | 1410 |
| LMArena WebDev | — | 1254 |
| 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), Qwen3.5 35B-A3B: —
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |
Reasoning Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 14.1 (#327), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| CritPt | 0% | 0.6% |
| LMArena Hard Prompts | 1257 | 1400 |
| DTBench | 59.5% | 80% |
| LMCA | 17.5% | 29.5% |
| Epoch Capabilities Index | 127.33 | 142.52 |
| SimpleBench | 19.9% | — |
| Chess Puzzles | — | 10% |
| LiveBench Reasoning | 50.8% | — |
| LiveBench Data Analysis | 49.5% | — |
| ForecastBench | 58.6 | — |
| LiveBench | 50.2% | — |
Math Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 15.3 (#298), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.1% | 70% |
| LMArena Math | 1267 | 1404 |
| MathArena Final-Answer Competitions | — | 56% |
| LiveBench Math | 42.2% | — |
| MATH Level 5 | 41.6% | — |
Knowledge Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 30.6 (#226), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| GPQA Diamond | 47.4% | 83.5% |
| Vectara Hallucination Rate | 4.1% | 10.5% |
| LMArena Expert | 1225 | 1408 |
| Confabulations | 22.8% | — |
| MMLU | 86.3% | — |
Multilingual Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 39.9 (#220), Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | 1236 | 1378 |
| LMArena Chinese | 1217 | 1457 |
| LMArena French | 1281 | 1412 |
| LMArena German | 1251 | 1367 |
| LMArena Japanese | 1150 | 1325 |
| LMArena Korean | 1143 | 1356 |
| LMArena Russian | 1252 | 1376 |
| LMArena Spanish | 1270 | 1392 |
Instruction Following Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 71.1 (#157), Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1242 | 1379 |
| LiveBench Instruction Following | 82.7% | — |
Long Context Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 26.4 (#295), Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1256 | 1389 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Qwen3.5 35B-A3B leads
Llama-3.3-70B-Instruct: 47.6 (#207), Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | Llama-3.3-70B-Instruct | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | 1274 | 1395 |
| LMArena Creative Writing | 1250 | 1346 |
| LMArena Multi-Turn | 1280 | 1390 |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 4.4× less per token, which makes it the better buy when Qwen3.5 35B-A3B's lead doesn't matter for your workload.
Which is cheaper, Llama-3.3-70B-Instruct or Qwen3.5 35B-A3B?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Qwen3.5 35B-A3B lists at $0.25 and $2.
Is Llama-3.3-70B-Instruct or Qwen3.5 35B-A3B better for coding?
Qwen3.5 35B-A3B scores higher on coding benchmarks: 33.8 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 35B-A3B does, with 262K tokens against 128K.
How many benchmarks do Llama-3.3-70B-Instruct and Qwen3.5 35B-A3B share?
25 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Qwen3.5 35B-A3B has 28.