Model comparison
Grok 4.5 vs Llama-3.3-70B-Instruct
Grok 4.5 is the stronger model overall, scoring 55.0 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 19× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
MetaSummary
- They share 26 benchmarks with published results for both. Grok 4.5 scores higher in 9 categories and Llama-3.3-70B-Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.5 leads 60.9 to 15.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 97.8% for Grok 4.5 and 5.1% for Llama-3.3-70B-Instruct.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $2 / $6 for Grok 4.5.
- Grok 4.5 accepts more context: 500K tokens versus 128K.
- Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.
Side by side
| Grok 4.5 | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | xAI | Meta |
| Noometry Index | 55.0 | 30.6 |
| Released | 2026-07-08 | 2024-12-06 |
| Weights | Proprietary | Open |
| Context window | 500K | 128K |
| Max output | 500K | 4K |
| Input $ / M tokens | $2 | $0.10 |
| Output $ / M tokens | $6 | $0.32 |
| Results tracked | 52 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.5 leads
Grok 4.5: 52.2 (#35), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| SciCode | 54.1% | 26% |
| WeirdML | 46.4% | 14.4% |
| LMArena Coding | 1474 | 1268 |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1553 | — |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
| ALE-Bench | 1,309 | — |
Agentic & Tool Use Grok 4.5 leads
Grok 4.5: 44.4 (#17), Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| APEX-Agents | 56.2% | — |
| Berkeley Function Calling Leaderboard | — | 31.9% |
| τ²-bench Banking | 47.9% | — |
| PostTrainBench | 23.4% | — |
| BALROG | — | 23% |
| GBAEval | 65.4% | — |
| GDP.pdf | 14% | — |
| LMArena Search | 1213 | — |
| Vending-Bench 2 | 3,887 | — |
Reasoning Grok 4.5 leads
Grok 4.5: 56.1 (#25), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| SimpleBench | 70% | 19.9% |
| CritPt | 15.4% | 0% |
| LMArena Hard Prompts | 1462 | 1257 |
| DTBench | 96.5% | 59.5% |
| LMCA | 45.2% | 17.5% |
| Epoch Capabilities Index | 153.92 | 127.33 |
| ARC-AGI-2 | 52.6% | — |
| Kagi LLM Benchmark | 83.5% | — |
| NYT Connections (extended) | 79.9% | — |
| ARC-AGI-1 | 87.2% | — |
| Chess Puzzles | 36% | — |
| LiveBench Reasoning | — | 50.8% |
| LiveBench Data Analysis | — | 49.5% |
| Surface Evolver Bench | 74.4% | — |
| ForecastBench | — | 58.6 |
| LiveBench | — | 50.2% |
Math Grok 4.5 leads
Grok 4.5: 60.9 (#35), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 97.8% | 5.1% |
| LMArena Math | 1459 | 1267 |
| FrontierMath (Tiers 1-3) | 57.2% | — |
| FrontierMath Tier 4 | 24.4% | — |
| ProofBench | 31% | — |
| LiveBench Math | — | 42.2% |
| MATH Level 5 | — | 41.6% |
Knowledge Grok 4.5 leads
Grok 4.5: 62.3 (#24), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 93.4% | 47.4% |
| LMArena Expert | 1466 | 1225 |
| SimpleQA Verified | 48.3% | — |
| Confabulations | — | 22.8% |
| Vectara Hallucination Rate | — | 4.1% |
| MMLU | — | 86.3% |
Multimodal Not comparable
Grok 4.5: 37.6 (#72), Llama-3.3-70B-Instruct: —
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Vision | 1288 | — |
| Blueprint-Bench 2 | 27.3% | — |
| Furniture Assembly | 22.5% | — |
| LMArena Document | 1452 | — |
Multilingual Grok 4.5 leads
Grok 4.5: 54.4 (#42), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1440 | 1236 |
| LMArena Chinese | 1496 | 1217 |
| LMArena French | 1456 | 1281 |
| LMArena German | 1446 | 1251 |
| LMArena Japanese | 1428 | 1150 |
| LMArena Korean | 1404 | 1143 |
| LMArena Russian | 1448 | 1252 |
| LMArena Spanish | 1450 | 1270 |
Instruction Following Grok 4.5 leads
Grok 4.5: 76.0 (#48), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1446 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context Grok 4.5 leads
Grok 4.5: 44.8 (#56), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1463 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Grok 4.5 leads
Grok 4.5: 65.8 (#42), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | Grok 4.5 | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1448 | 1274 |
| LMArena Creative Writing | 1442 | 1250 |
| LMArena Multi-Turn | 1456 | 1280 |
| EQ-Bench Creative Writing | 1579 | — |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is Grok 4.5 better than Llama-3.3-70B-Instruct?
Grok 4.5 is the stronger model overall, scoring 55.0 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 19× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Which is cheaper, Grok 4.5 or Llama-3.3-70B-Instruct?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Grok 4.5 lists at $2 and $6.
Is Grok 4.5 or Llama-3.3-70B-Instruct better for coding?
Grok 4.5 scores higher on coding benchmarks: 52.2 versus 31.0 in the Noometry coding category.
Which has the bigger context window?
Grok 4.5 does, with 500K tokens against 128K.
How many benchmarks do Grok 4.5 and Llama-3.3-70B-Instruct share?
26 benchmarks have published results for both models. Grok 4.5 has 52 scored results on Noometry and Llama-3.3-70B-Instruct has 43.