Noometry

Model comparison

Llama 2-13B vs MiniMax-M2.7

MiniMax-M2.7 is the stronger model overall, scoring 37.7 to 29.6 on the Noometry Index.

Last verified . 18 shared benchmarks.

Llama 2-13B Meta

29.6

Rank #309 Confirmed

MiniMax-M2.7 MiniMax

37.7

Rank #196 Confirmed

Summary

  • They share 18 benchmarks with published results for both. Llama 2-13B scores higher in 1 category and MiniMax-M2.7 in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where MiniMax-M2.7 leads 58.9 to 29.8.

Side by side

Llama 2-13BMiniMax-M2.7
ProviderMetaMiniMax
Noometry Index29.637.7
Released2023-07-182026-03-18
WeightsOpenOpen
Context window—205K
Max output—131K
Input $ / M tokens—$0.30
Output $ / M tokens—$1.20
Results tracked3230

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding MiniMax-M2.7 leads

Llama 2-13B: 30.9 (#291), MiniMax-M2.7: 41.8 (#120)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Coding10621454
LMArena WebDev—1398
SciCode—47%
WeirdML—37%
ALE-Bench—599.25

Agentic & Tool Use Not comparable

Llama 2-13B: —, MiniMax-M2.7: 25.1 (#111)

BenchmarkLlama 2-13BMiniMax-M2.7
Terminal-Bench—45.1%
ExploitBench—13.3%
GBAEval—0%

Reasoning MiniMax-M2.7 leads

Llama 2-13B: 12.8 (#337), MiniMax-M2.7: 19.7 (#253)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Hard Prompts10511422
Epoch Capabilities Index106.17145.85
NYT Connections (extended)—24.7%
CritPt—0.6%
Chess Puzzles0%—
Thematic Generalization—39.3%
DTBench42.2%—
BIG-Bench Hard58.2%—
HellaSwag80.7%—
LAMBADA76.5%—
PIQA80.8%—
WinoGrande72.8%—

Math Llama 2-13B leads

Llama 2-13B: 31.1 (#229), MiniMax-M2.7: 25.9 (#263)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Math10651420
ProofBench—3%
GSM8K36.9%—

Knowledge MiniMax-M2.7 leads

Llama 2-13B: 28.1 (#249), MiniMax-M2.7: 37.7 (#152)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Expert10301444
Vectara Hallucination Rate—12.9%
ARC (AI2) Challenge60.3%—
BoolQ82.4%—
MMLU55.6%—
OpenBookQA57%—
TriviaQA79.6%—

Multimodal Not comparable

Llama 2-13B: —, MiniMax-M2.7: —

BenchmarkLlama 2-13BMiniMax-M2.7
ScienceQA55.8%—

Multilingual MiniMax-M2.7 leads

Llama 2-13B: 26.5 (#279), MiniMax-M2.7: 50.3 (#123)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Non-English10241382
LMArena Chinese10011441
LMArena French10441421
LMArena German10091398
LMArena Japanese8941262
LMArena Korean9531313
LMArena Russian10551383
LMArena Spanish10871403

Instruction Following MiniMax-M2.7 leads

Llama 2-13B: 53.3 (#287), MiniMax-M2.7: 74.1 (#103)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Instruction Following10451405

Long Context MiniMax-M2.7 leads

Llama 2-13B: 32.3 (#269), MiniMax-M2.7: 43.3 (#99)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Longer Query10641419

Writing & Preference MiniMax-M2.7 leads

Llama 2-13B: 29.8 (#289), MiniMax-M2.7: 58.9 (#112)

BenchmarkLlama 2-13BMiniMax-M2.7
LMArena Text10841405
LMArena Creative Writing10471354
LMArena Multi-Turn10501412

Frequently asked questions

Is Llama 2-13B better than MiniMax-M2.7?

MiniMax-M2.7 is the stronger model overall, scoring 37.7 to 29.6 on the Noometry Index.

Is Llama 2-13B or MiniMax-M2.7 better for coding?

MiniMax-M2.7 scores higher on coding benchmarks: 41.8 versus 30.9 in the Noometry coding category.

How many benchmarks do Llama 2-13B and MiniMax-M2.7 share?

18 benchmarks have published results for both models. Llama 2-13B has 32 scored results on Noometry and MiniMax-M2.7 has 30.

Related comparisons

Go deeper