Llama 3.3 70B Instruct Turbo

Llama 3.3 70B Instruct Turbo is a model from Meta. It has 5 scored results in 3 categories. It has no index.

  • Input: text
UnrankedEnough results for an index.
IndexAll categories, out of 100.UnrankedMedian 51.0
CoverageIndex weight that has results.45%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.—Median 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.None counted——
CodingCode writing and repair.1 counted19.5—
ReasoningLogic problems and puzzles.None counted——
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.2 counted27.8—
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.2 counted22.1—

Results

5 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH 500Math73.4% ±3.947 / 59
AIMEMath16.0% ±1.183 / 95
MMLU ProKnowledge69.9% ±0.4118 / 135
GPQA Diamond · Vals AIKnowledge50.0% ±2.5121 / 135
LiveCodeBenchCoding36.3% ±1.0134 / 142
MGSMMultilingual91.1% ±1.830 / 74
LegalBenchLegal77.2% ±0.8110 / 149
TaxEval v2Finance59.4% ±1.0131 / 143

Sources

BenchLMCC BY-NC 4.0Vals AINo licence stated

More from Meta

All models
Muse Wonder Full77.3Muse Wonder73.1Muse Spark 1.370.1Muse Spark Open65.1Muse Spark 1.264.9