Model comparison
Codellama 34b Instruct vs MiniMax-M2.5
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 30.8 on the Noometry Index.
Last verified . 10 shared benchmarks.
Meta
MiniMaxSummary
- They share 10 benchmarks with published results for both. Codellama 34b Instruct scores higher in 2 categories and MiniMax-M2.5 in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.5 leads 53.9 to 28.2.
Side by side
| Codellama 34b Instruct | MiniMax-M2.5 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 30.8 | 38.3 |
| Released | — | 2026-02-12 |
| Weights | Open | Open |
| Context window | — | 205K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 14 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Codellama 34b Instruct: 28.5 (#314), MiniMax-M2.5: 48.1 (#58)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1046 | 1381 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| BigCodeBench Instruct | 29% | — |
| BigCodeBench Complete | 37.1% | — |
| ALE-Bench | — | 618.17 |
| HumanEval+ | 43.9% | — |
| MBPP+ | 56.3% | — |
Agentic & Tool Use Not comparable
Codellama 34b Instruct: —, MiniMax-M2.5: 30.4 (#77)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |
Reasoning Codellama 34b Instruct leads
Codellama 34b Instruct: 19.6 (#255), MiniMax-M2.5: 17.5 (#292)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Hard Prompts | 1032 | 1372 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 55.2% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| Epoch Capabilities Index | — | 146.68 |
Math Codellama 34b Instruct leads
Codellama 34b Instruct: 31.0 (#230), MiniMax-M2.5: 26.9 (#253)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1056 | 1378 |
| ProofBench | — | 4% |
Knowledge Not comparable
Codellama 34b Instruct: —, MiniMax-M2.5: 39.2 (#135)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | — | 9.1% |
| LMArena Expert | — | 1379 |
Multilingual MiniMax-M2.5 leads
Codellama 34b Instruct: 25.8 (#284), MiniMax-M2.5: 47.1 (#152)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1011 | 1338 |
| LMArena Chinese | 976 | 1393 |
| LMArena French | — | 1362 |
| LMArena German | — | 1362 |
| LMArena Japanese | — | 1171 |
| LMArena Korean | — | 1232 |
| LMArena Russian | — | 1358 |
| LMArena Spanish | — | 1354 |
Instruction Following MiniMax-M2.5 leads
Codellama 34b Instruct: 52.2 (#291), MiniMax-M2.5: 71.5 (#148)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1028 | 1353 |
Long Context MiniMax-M2.5 leads
Codellama 34b Instruct: 30.9 (#284), MiniMax-M2.5: 37.5 (#216)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1013 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference MiniMax-M2.5 leads
Codellama 34b Instruct: 28.2 (#297), MiniMax-M2.5: 53.9 (#153)
| Benchmark | Codellama 34b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1066 | 1359 |
| LMArena Creative Writing | 1032 | 1331 |
| LMArena Multi-Turn | 1015 | 1364 |
| EQ-Bench Creative Writing | — | 1361 |
Frequently asked questions
Is Codellama 34b Instruct better than MiniMax-M2.5?
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 30.8 on the Noometry Index.
Is Codellama 34b Instruct or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 28.5 in the Noometry coding category.
How many benchmarks do Codellama 34b Instruct and MiniMax-M2.5 share?
10 benchmarks have published results for both models. Codellama 34b Instruct has 14 scored results on Noometry and MiniMax-M2.5 has 33.