Model comparison
Codellama 70b Instruct vs GPT-6 Luna
GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.7 on the Noometry Index.
Last verified . 4 shared benchmarks.
MetaSummary
- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 0 categories and GPT-6 Luna in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Luna leads 48.2 to 20.1.
- Codellama 70b Instruct has downloadable open weights; the other is API-only.
Side by side
| Codellama 70b Instruct | GPT-6 Luna | |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 33.7 | 53.3 |
| Released | — | 2026-09-22 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.50 |
| Results tracked | 7 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
Codellama 70b Instruct: 37.6 (#193), GPT-6 Luna: 55.5 (#25)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
| LMArena WebDev | — | 1581 |
| SciCode | — | 54.6% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1439 |
| BigCodeBench Complete | 49.6% | — |
| ALE-Bench | — | 1,577 |
| HumanEval+ | 65.9% | — |
Agentic & Tool Use Not comparable
Codellama 70b Instruct: —, GPT-6 Luna: 33.3 (#54)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| APEX-Agents | — | 44.3% |
| GDP.pdf | — | 23% |
Reasoning GPT-6 Luna leads
Codellama 70b Instruct: 20.1 (#242), GPT-6 Luna: 48.2 (#41)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1411 |
| ARC-AGI-2 | — | 59.3% |
| NYT Connections (extended) | — | 68.7% |
| ARC-AGI-1 | — | 86.7% |
| CritPt | — | 19.4% |
| Chess Puzzles | — | 31% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 90.1% |
| LMCA | — | 44.5% |
| Epoch Capabilities Index | — | 156.28 |
Math Not comparable
Codellama 70b Instruct: —, GPT-6 Luna: 76.1 (#15)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 78.9% |
| FrontierMath Tier 4 | — | 56.1% |
| OTIS Mock AIME 2024-2025 | — | 98.9% |
| ProofBench | — | 64% |
| LMArena Math | — | 1416 |
Knowledge Not comparable
Codellama 70b Instruct: —, GPT-6 Luna: 57.0 (#41)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | — | 90.5% |
| SimpleQA Verified | — | 41.4% |
| LMArena Expert | — | 1444 |
Multimodal Not comparable
Codellama 70b Instruct: —, GPT-6 Luna: 42.4 (#30)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| LMArena Vision | — | 1217 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 44.2% |
Multilingual GPT-6 Luna leads
Codellama 70b Instruct: 24.8 (#288), GPT-6 Luna: 50.5 (#117)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 992 | 1386 |
| LMArena Chinese | — | 1433 |
| LMArena French | — | 1420 |
| LMArena German | — | 1369 |
| LMArena Japanese | — | 1369 |
| LMArena Korean | — | 1360 |
| LMArena Russian | — | 1394 |
| LMArena Spanish | — | 1393 |
Instruction Following GPT-6 Luna leads
Codellama 70b Instruct: 51.9 (#293), GPT-6 Luna: 74.3 (#99)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1024 | 1409 |
Long Context Not comparable
Codellama 70b Instruct: —, GPT-6 Luna: 43.0 (#111)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | — | 1409 |
Writing & Preference GPT-6 Luna leads
Codellama 70b Instruct: 33.4 (#277), GPT-6 Luna: 58.3 (#119)
| Benchmark | Codellama 70b Instruct | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1057 | 1391 |
| LMArena Creative Writing | — | 1363 |
| LMArena Multi-Turn | — | 1396 |
Frequently asked questions
Is Codellama 70b Instruct better than GPT-6 Luna?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.7 on the Noometry Index.
Is Codellama 70b Instruct or GPT-6 Luna better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 37.6 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and GPT-6 Luna share?
4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and GPT-6 Luna has 42.