Model comparison
Llama-3.3-70B-Instruct vs MiniMax-M2.5
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 30.3 on the Noometry Index. Llama-3.3-70B-Instruct costs 3.4× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Meta
MiniMaxSummary
- They share 19 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 0 categories and MiniMax-M2.5 in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 47.9 to 30.6.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 4.1% for Llama-3.3-70B-Instruct and 9.1% for MiniMax-M2.5.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- MiniMax-M2.5 accepts more context: 205K tokens versus 128K.
Side by side
| Llama-3.3-70B-Instruct | MiniMax-M2.5 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 30.3 | 38.1 |
| Released | 2024-12-06 | 2026-02-12 |
| Weights | Open | Open |
| Context window | 128K | 205K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.10 | $0.30 |
| Output $ / M tokens | $0.32 | $1.20 |
| Results tracked | 43 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 30.6 (#284), MiniMax-M2.5: 47.9 (#53)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1268 | 1380 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 26% | — |
| WeirdML | 14.4% | — |
| BigCodeBench Instruct | 46.9% | — |
| LiveBench Coding | 36.6% | — |
| BigCodeBench Complete | 57.5% | — |
| ALE-Bench | — | 618.17 |
Agentic & Tool Use MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 25.6 (#100), MiniMax-M2.5: 30.2 (#70)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |
| Vending-Bench 2 | — | -23.16 |
Reasoning MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 13.7 (#325), MiniMax-M2.5: 17.6 (#281)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1372 |
| Epoch Capabilities Index | 127.32 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | 19.9% | — |
| Kagi LLM Benchmark | — | 55.2% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | 0% | — |
| LiveBench Reasoning | 50.8% | — |
| DTBench | 59.5% | — |
| LiveBench Data Analysis | 49.5% | — |
| LMCA | 17.5% | — |
| ForecastBench | 58.6 | — |
| LiveBench | 50.2% | — |
Math MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 14.3 (#292), MiniMax-M2.5: 25.8 (#242)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1268 | 1378 |
| OTIS Mock AIME 2024-2025 | 5.1% | — |
| ProofBench | — | 4% |
| LiveBench Math | 42.2% | — |
| MATH Level 5 | 41.6% | — |
Knowledge MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 30.5 (#220), MiniMax-M2.5: 38.6 (#132)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 4.1% | 9.1% |
| LMArena Expert | 1225 | 1378 |
| GPQA Diamond | 47.4% | — |
| Confabulations | 22.8% | — |
| MMLU | 86.3% | — |
Multilingual MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 40.1 (#215), MiniMax-M2.5: 47.3 (#147)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1236 | 1338 |
| LMArena Chinese | 1217 | 1392 |
| LMArena French | 1280 | 1363 |
| LMArena German | 1251 | 1361 |
| LMArena Japanese | 1149 | 1169 |
| LMArena Korean | 1142 | 1232 |
| LMArena Russian | 1252 | 1359 |
| LMArena Spanish | 1269 | 1354 |
Instruction Following Too close to call
Llama-3.3-70B-Instruct: 71.3 (#152), MiniMax-M2.5: 71.7 (#143)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1242 | 1353 |
| LiveBench Instruction Following | 82.7% | — |
Long Context MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 26.4 (#290), MiniMax-M2.5: 37.4 (#211)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1256 | 1367 |
| Fiction.LiveBench | 33.3% | — |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference MiniMax-M2.5 leads
Llama-3.3-70B-Instruct: 47.8 (#203), MiniMax-M2.5: 54.1 (#149)
| Benchmark | Llama-3.3-70B-Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1274 | 1359 |
| LMArena Creative Writing | 1250 | 1331 |
| LMArena Multi-Turn | 1279 | 1364 |
| EQ-Bench Creative Writing | — | 1361 |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than MiniMax-M2.5?
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 30.3 on the Noometry Index. Llama-3.3-70B-Instruct costs 3.4× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Which is cheaper, Llama-3.3-70B-Instruct or MiniMax-M2.5?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is Llama-3.3-70B-Instruct or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 30.6 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 128K.
How many benchmarks do Llama-3.3-70B-Instruct and MiniMax-M2.5 share?
19 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and MiniMax-M2.5 has 33.