Model comparison
Llama 3.1-8B vs MiniMax-M2
MiniMax-M2 is the stronger model overall, scoring 37.4 to 23.0 on the Noometry Index. Llama 3.1-8B costs 9.1× less per token, which makes it the better buy when MiniMax-M2's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Meta
MiniMaxSummary
- They share 15 benchmarks with published results for both. Llama 3.1-8B scores higher in 0 categories and MiniMax-M2 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2 leads 37.0 to 8.0.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.
- MiniMax-M2 accepts more context: 205K tokens versus 128K.
Side by side
| Llama 3.1-8B | MiniMax-M2 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 23.0 | 37.4 |
| Released | 2024-07-23 | 2025-10-27 |
| Weights | Open | Open |
| Context window | 128K | 205K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.05 | $0.30 |
| Output $ / M tokens | $0.08 | $1.20 |
| Results tracked | 43 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2 leads
Llama 3.1-8B: 20.2 (#340), MiniMax-M2: 39.3 (#159)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Coding | 1195 | 1370 |
| SWE-bench Verified (bash only) | — | 61% |
| LMArena WebDev | — | 1297 |
| SciCode | 13.2% | — |
| WeirdML | 1.7% | — |
| BigCodeBench Instruct | 32.8% | — |
| BigCodeBench Complete | 40.5% | — |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use MiniMax-M2 leads
Llama 3.1-8B: 22.5 (#131), MiniMax-M2: 25.1 (#109)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| Terminal-Bench | — | 30% |
| Berkeley Function Calling Leaderboard | 25.8% | — |
| BALROG | 15.1% | — |
| Vending-Bench 2 | — | 160.6 |
Reasoning MiniMax-M2 leads
Llama 3.1-8B: 14.9 (#321), MiniMax-M2: 19.4 (#258)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Hard Prompts | 1175 | 1357 |
| Kagi LLM Benchmark | — | 57.8% |
| NYT Connections (extended) | — | 14.8% |
| CritPt | 0% | — |
| Chess Puzzles | 0% | — |
| DTBench | 50.9% | — |
| LMCA | 5.4% | — |
| Epoch Capabilities Index | 116.57 | — |
| PIQA | 81.2% | — |
Math MiniMax-M2 leads
Llama 3.1-8B: 10.2 (#317), MiniMax-M2: 37.3 (#160)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Math | 1179 | 1352 |
| OTIS Mock AIME 2024-2025 | 1.7% | — |
| Omni-MATH | 13.7% | — |
| MATH Level 5 | 22.9% | — |
| GSM8K | 82.4% | — |
Knowledge MiniMax-M2 leads
Llama 3.1-8B: 8.0 (#307), MiniMax-M2: 37.0 (#163)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Expert | 1144 | 1337 |
| GPQA Diamond | 27% | — |
| MMLU-Pro | 40.6% | — |
| GPQA (HELM) | 24.7% | — |
| BoolQ | 82.8% | — |
| MMLU | 56.1% | — |
Multilingual MiniMax-M2 leads
Llama 3.1-8B: 34.0 (#249), MiniMax-M2: 45.3 (#171)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Non-English | 1148 | 1313 |
| LMArena Chinese | 1151 | 1366 |
| LMArena French | 1177 | 1335 |
| LMArena German | 1144 | 1355 |
| LMArena Russian | 1158 | 1331 |
| LMArena Spanish | 1169 | 1326 |
| LMArena Japanese | 1061 | — |
| LMArena Korean | 1053 | — |
Instruction Following MiniMax-M2 leads
Llama 3.1-8B: 58.9 (#258), MiniMax-M2: 70.2 (#166)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Instruction Following | 1159 | 1328 |
| IFEval | 74.3% | — |
Long Context MiniMax-M2 leads
Llama 3.1-8B: 35.8 (#238), MiniMax-M2: 40.5 (#153)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Longer Query | 1182 | 1331 |
Writing & Preference MiniMax-M2 leads
Llama 3.1-8B: 29.7 (#290), MiniMax-M2: 53.0 (#162)
| Benchmark | Llama 3.1-8B | MiniMax-M2 |
|---|---|---|
| LMArena Text | 1187 | 1340 |
| LMArena Creative Writing | 1154 | 1286 |
| LMArena Multi-Turn | 1172 | 1361 |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
Frequently asked questions
Is Llama 3.1-8B better than MiniMax-M2?
MiniMax-M2 is the stronger model overall, scoring 37.4 to 23.0 on the Noometry Index. Llama 3.1-8B costs 9.1× less per token, which makes it the better buy when MiniMax-M2's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-8B or MiniMax-M2?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; MiniMax-M2 lists at $0.30 and $1.20.
Is Llama 3.1-8B or MiniMax-M2 better for coding?
MiniMax-M2 scores higher on coding benchmarks: 39.3 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2 does, with 205K tokens against 128K.
How many benchmarks do Llama 3.1-8B and MiniMax-M2 share?
15 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and MiniMax-M2 has 21.