Model comparison
Codellama 34b Instruct vs GPT-4o mini
Codellama 34b Instruct is the stronger model overall, scoring 30.2 to 25.4 on the Noometry Index.
Last verified . 14 shared benchmarks.
MetaSummary
- They share 14 benchmarks with published results for both. Codellama 34b Instruct scores higher in 3 categories and GPT-4o mini in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Codellama 34b Instruct leads 28.5 to 9.9.
- The biggest single-benchmark swing is BigCodeBench Complete: 37.1% for Codellama 34b Instruct and 57.4% for GPT-4o mini.
- Codellama 34b Instruct has downloadable open weights; the other is API-only.
Side by side
| Codellama 34b Instruct | GPT-4o mini | |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 30.2 | 25.4 |
| Released | — | 2024-07-18 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 14 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Codellama 34b Instruct leads
Codellama 34b Instruct: 28.0 (#310), GPT-4o mini: 21.6 (#330)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| BigCodeBench Instruct | 29% | 46.1% |
| LMArena Coding | 1046 | 1290 |
| BigCodeBench Complete | 37.1% | 57.4% |
| HumanEval+ | 43.9% | 83.5% |
| MBPP+ | 56.3% | 72.2% |
| Aider Polyglot | — | 3.6% |
| WeirdML | — | 11.8% |
| LiveBench Coding | — | 43.1% |
Agentic & Tool Use Not comparable
Codellama 34b Instruct: —, GPT-4o mini: 27.2 (#95)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| BALROG | — | 17.4% |
Reasoning Codellama 34b Instruct leads
Codellama 34b Instruct: 18.8 (#250), GPT-4o mini: 8.7 (#343)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Hard Prompts | 1032 | 1267 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 10.7% |
| Kagi LLM Benchmark | — | 28.8% |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 32.8% |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 54.4% |
| LiveBench Data Analysis | — | 50% |
| LMCA | — | 10.4% |
| Epoch Capabilities Index | — | 126.56 |
| LiveBench | — | 41.3% |
| PIQA | — | 88.7% |
Math Codellama 34b Instruct leads
Codellama 34b Instruct: 28.5 (#227), GPT-4o mini: 9.9 (#307)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Math | 1056 | 1267 |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.9% |
| Omni-MATH | — | 28% |
| LiveBench Math | — | 36.3% |
| MATH Level 5 | — | 52.6% |
| GSM8K | — | 91.3% |
Knowledge Not comparable
Codellama 34b Instruct: —, GPT-4o mini: 17.7 (#279)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| GPQA Diamond | — | 37.7% |
| SimpleQA Verified | — | 8.3% |
| MMLU-Pro | — | 60.3% |
| Confabulations | — | 37.2% |
| GPQA (HELM) | — | 36.8% |
| LMArena Expert | — | 1235 |
| BoolQ | — | 88.7% |
| MMLU | — | 81.8% |
Multimodal Not comparable
Codellama 34b Instruct: —, GPT-4o mini: 25.3 (#117)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Vision | — | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |
Multilingual GPT-4o mini leads
Codellama 34b Instruct: 25.9 (#279), GPT-4o mini: 42.2 (#194)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1011 | 1266 |
| LMArena Chinese | 976 | 1264 |
| LMArena French | — | 1296 |
| LMArena German | — | 1272 |
| LMArena Japanese | — | 1216 |
| LMArena Korean | — | 1195 |
| LMArena Russian | — | 1275 |
| LMArena Spanish | — | 1276 |
Instruction Following GPT-4o mini leads
Codellama 34b Instruct: 52.4 (#286), GPT-4o mini: 62.1 (#234)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Instruction Following | 1028 | 1259 |
| LiveBench Instruction Following | — | 56.8% |
| IFEval | — | 78.2% |
Long Context GPT-4o mini leads
Codellama 34b Instruct: 30.8 (#279), GPT-4o mini: 39.1 (#181)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1014 | 1289 |
Writing & Preference GPT-4o mini leads
Codellama 34b Instruct: 28.4 (#292), GPT-4o mini: 39.7 (#243)
| Benchmark | Codellama 34b Instruct | GPT-4o mini |
|---|---|---|
| LMArena Text | 1066 | 1286 |
| LMArena Creative Writing | 1032 | 1268 |
| LMArena Multi-Turn | 1015 | 1285 |
| Short-Story Creative Writing | — | 67.2% |
| EQ-Bench Creative Writing | — | 873 |
| WildBench | — | 79.1% |
| LiveBench Language | — | 28.6% |
Frequently asked questions
Is Codellama 34b Instruct better than GPT-4o mini?
Codellama 34b Instruct is the stronger model overall, scoring 30.2 to 25.4 on the Noometry Index.
Is Codellama 34b Instruct or GPT-4o mini better for coding?
Codellama 34b Instruct scores higher on coding benchmarks: 28.0 versus 21.6 in the Noometry coding category.
How many benchmarks do Codellama 34b Instruct and GPT-4o mini share?
14 benchmarks have published results for both models. Codellama 34b Instruct has 14 scored results on Noometry and GPT-4o mini has 60.