Model comparison
Codellama 34b Instruct vs GPT-5.4 nano
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 30.8 on the Noometry Index.
Last verified . 10 shared benchmarks.
MetaSummary
- They share 10 benchmarks with published results for both. Codellama 34b Instruct scores higher in 0 categories and GPT-5.4 nano in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-5.4 nano leads 55.7 to 28.2.
- Codellama 34b Instruct has downloadable open weights; the other is API-only.
Side by side
| Codellama 34b Instruct | GPT-5.4 nano | |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 30.8 | 41.9 |
| Released | — | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.20 |
| Output $ / M tokens | — | $1.25 |
| Results tracked | 14 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
Codellama 34b Instruct: 28.5 (#314), GPT-5.4 nano: 43.6 (#84)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Coding | 1046 | 1405 |
| SciCode | — | 46.9% |
| WeirdML | — | 49.2% |
| BigCodeBench Instruct | 29% | — |
| BigCodeBench Complete | 37.1% | — |
| ALE-Bench | — | 1,005 |
| HumanEval+ | 43.9% | — |
| MBPP+ | 56.3% | — |
Reasoning GPT-5.4 nano leads
Codellama 34b Instruct: 19.6 (#255), GPT-5.4 nano: 23.7 (#173)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Hard Prompts | 1032 | 1381 |
| ARC-AGI-2 | — | 5.7% |
| Kagi LLM Benchmark | — | 39.7% |
| ARC-AGI-1 | — | 51.5% |
| CritPt | — | 9.3% |
| Chess Puzzles | — | 30% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 36.9% |
| Epoch Capabilities Index | — | 145.81 |
| ForecastBench | — | 57.3 |
Math GPT-5.4 nano leads
Codellama 34b Instruct: 31.0 (#230), GPT-5.4 nano: 40.9 (#88)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Math | 1056 | 1406 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| OTIS Mock AIME 2024-2025 | — | 87.8% |
| ProofBench | — | 5% |
| FrontierMath (Feb 2025 set) | — | 25.9% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge Not comparable
Codellama 34b Instruct: —, GPT-5.4 nano: 41.9 (#103)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | — | 78.5% |
| SimpleQA Verified | — | 11.7% |
| Vectara Hallucination Rate | — | 3.1% |
| LMArena Expert | — | 1396 |
Multimodal Not comparable
Codellama 34b Instruct: —, GPT-5.4 nano: 36.7 (#78)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | — | 1196 |
Multilingual GPT-5.4 nano leads
Codellama 34b Instruct: 25.8 (#284), GPT-5.4 nano: 48.6 (#140)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1011 | 1359 |
| LMArena Chinese | 976 | 1392 |
| LMArena French | — | 1396 |
| LMArena German | — | 1367 |
| LMArena Japanese | — | 1343 |
| LMArena Korean | — | 1320 |
| LMArena Russian | — | 1363 |
| LMArena Spanish | — | 1371 |
Instruction Following GPT-5.4 nano leads
Codellama 34b Instruct: 52.2 (#291), GPT-5.4 nano: 71.9 (#144)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1028 | 1362 |
Long Context GPT-5.4 nano leads
Codellama 34b Instruct: 30.9 (#284), GPT-5.4 nano: 41.6 (#137)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1013 | 1366 |
Writing & Preference GPT-5.4 nano leads
Codellama 34b Instruct: 28.2 (#297), GPT-5.4 nano: 55.7 (#142)
| Benchmark | Codellama 34b Instruct | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1066 | 1372 |
| LMArena Creative Writing | 1032 | 1314 |
| LMArena Multi-Turn | 1015 | 1382 |
Frequently asked questions
Is Codellama 34b Instruct better than GPT-5.4 nano?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 30.8 on the Noometry Index.
Is Codellama 34b Instruct or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 28.5 in the Noometry coding category.
How many benchmarks do Codellama 34b Instruct and GPT-5.4 nano share?
10 benchmarks have published results for both models. Codellama 34b Instruct has 14 scored results on Noometry and GPT-5.4 nano has 40.