Model comparison
GPT-5.6 Terra vs Llama 4 Scout
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 27.7 on the Noometry Index. Llama 4 Scout costs 30× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
MetaSummary
- They share 29 benchmarks with published results for both. GPT-5.6 Terra scores higher in 10 categories and Llama 4 Scout in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 19.6.
- The biggest single-benchmark swing is ARC-AGI-1: 96.5% for GPT-5.6 Terra and 0.5% for Llama 4 Scout.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 128K.
- Llama 4 Scout has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Terra | Llama 4 Scout | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 59.2 | 27.7 |
| Released | 2026-07-09 | 2025-04-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $2 | $0.10 |
| Output $ / M tokens | $12 | $0.30 |
| Results tracked | 52 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
GPT-5.6 Terra: 57.7 (#19), Llama 4 Scout: 20.2 (#339)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| SciCode | 55% | 17% |
| LMArena Coding | 1484 | 1286 |
| DeepSWE | 69.6% | — |
| FrontierCode | 41.3% | — |
| SWE-bench Verified (bash only) | — | 9.1% |
| CursorBench | 41.3% | — |
| LMArena WebDev | 1522 | — |
| WeirdML | 78.3% | — |
| BigCodeBench Complete | — | 43.1% |
| ALE-Bench | 1,951 | — |
Agentic & Tool Use GPT-5.6 Terra leads
GPT-5.6 Terra: 40.1 (#25), Llama 4 Scout: 24.6 (#119)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| APEX-Agents | 58.2% | — |
| Berkeley Function Calling Leaderboard | — | 28.1% |
| BALROG | 53.2% | — |
| GDP.pdf | 24.7% | — |
| Vending-Bench 2 | 7,343 | — |
Reasoning GPT-5.6 Terra leads
GPT-5.6 Terra: 60.7 (#21), Llama 4 Scout: 9.1 (#345)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| ARC-AGI-2 | 83.9% | 0% |
| Kagi LLM Benchmark | 51.3% | 36.9% |
| ARC-AGI-1 | 96.5% | 0.5% |
| CritPt | 30% | 0% |
| LMArena Hard Prompts | 1468 | 1266 |
| DTBench | 93.3% | 57.9% |
| LMCA | 55% | 12% |
| Epoch Capabilities Index | 159.62 | 129.64 |
| SimpleBench | 48.9% | — |
| NYT Connections (extended) | 78.4% | — |
| Chess Puzzles | 54% | — |
| Mystery Game Puzzles | 35% | — |
| Surface Evolver Bench | 83.8% | — |
| ForecastBench | — | 57.5 |
Math GPT-5.6 Terra leads
GPT-5.6 Terra: 81.6 (#12), Llama 4 Scout: 19.6 (#286)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.7% | 7.8% |
| LMArena Math | 1466 | 1287 |
| FrontierMath (Tiers 1-3) | 86% | — |
| FrontierMath Tier 4 | 70.7% | — |
| ProofBench | 74% | — |
| Omni-MATH | — | 37.3% |
| MATH Level 5 | — | 62.3% |
| FrontierMath (Feb 2025 set) | — | 0% |
Knowledge GPT-5.6 Terra leads
GPT-5.6 Terra: 61.2 (#30), Llama 4 Scout: 31.9 (#217)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| GPQA Diamond | 93.3% | 51.8% |
| LMArena Expert | 1492 | 1235 |
| SimpleQA Verified | 43.2% | — |
| MMLU-Pro | — | 74.2% |
| Vectara Hallucination Rate | — | 7.7% |
| GPQA (HELM) | — | 50.7% |
Multimodal GPT-5.6 Terra leads
GPT-5.6 Terra: 47.3 (#11), Llama 4 Scout: 32.2 (#102)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| LMArena Vision | 1271 | 1118 |
| Blueprint-Bench 2 | 30.8% | — |
| Furniture Assembly | 54.2% | — |
| LMArena Document | 1472 | — |
| SpatialViz-Bench | — | 34.2% |
Multilingual GPT-5.6 Terra leads
GPT-5.6 Terra: 54.4 (#44), Llama 4 Scout: 41.0 (#212)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| LMArena Non-English | 1439 | 1252 |
| LMArena Chinese | 1513 | 1255 |
| LMArena French | 1471 | 1282 |
| LMArena German | 1460 | 1272 |
| LMArena Japanese | 1457 | 1206 |
| LMArena Korean | 1425 | 1207 |
| LMArena Russian | 1450 | 1263 |
| LMArena Spanish | 1448 | 1278 |
Instruction Following GPT-5.6 Terra leads
GPT-5.6 Terra: 76.4 (#40), Llama 4 Scout: 65.8 (#217)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| LMArena Instruction Following | 1454 | 1248 |
| IFEval | — | 81.8% |
Long Context GPT-5.6 Terra leads
GPT-5.6 Terra: 44.4 (#68), Llama 4 Scout: 27.5 (#294)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| LMArena Longer Query | 1451 | 1265 |
| Fiction.LiveBench | — | 36% |
Writing & Preference GPT-5.6 Terra leads
GPT-5.6 Terra: 70.2 (#23), Llama 4 Scout: 37.0 (#261)
| Benchmark | GPT-5.6 Terra | Llama 4 Scout |
|---|---|---|
| LMArena Text | 1447 | 1279 |
| LMArena Creative Writing | 1410 | 1249 |
| EQ-Bench Creative Writing | 1855 | 783 |
| LMArena Multi-Turn | 1449 | 1280 |
| WildBench | — | 78% |
| EQ-Bench 4 | 1234 | — |
Frequently asked questions
Is GPT-5.6 Terra better than Llama 4 Scout?
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 27.7 on the Noometry Index. Llama 4 Scout costs 30× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Which is cheaper, GPT-5.6 Terra or Llama 4 Scout?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GPT-5.6 Terra lists at $2 and $12.
Is GPT-5.6 Terra or Llama 4 Scout better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Terra does, with 1.05M tokens against 128K.
How many benchmarks do GPT-5.6 Terra and Llama 4 Scout share?
29 benchmarks have published results for both models. GPT-5.6 Terra has 52 scored results on Noometry and Llama 4 Scout has 43.