Skip to main content

Llama 3.1 Nemotron Instruct 70B

NVIDIALlamaOpen WeightLlama 3.1 Community License

Description

A large language model customized by NVIDIA to improve the helpfulness of LLM generated responses. It is a fine-tuned version of Llama 3.1 70B Instruct. The model was trained using RLHF (REINFORCE) with HelpSteer2-Preference prompts.

Release Date
2024-10-15
Parameters
70.0B
Context Length
Modalities

Capability Radar

26
general
18
coding
27
reasoning
30
scienceest.
24
agents
0
multimodal

Science uses a reasoning proxy when dedicated science benchmarks are unavailable.

Rankings

Domain#RankScoreSource
Code Ranking436
11.0
AA
General Ranking378
29.0
AA
Math Reasoning282
26.0
AA
Reasoning18
86.0
LS
Science373
30.0
AA

Benchmark Scores (LLM Stats)

Communication

MT-Bench0.09 / 100SR

Finance

MMLU Chat80.6%SR
MMLU80.2%SR
TruthfulQA58.6%SR

General

Instruct HumanEval73.8%SR
ARC-C69.2%SR

Language

Winogrande84.5%SR
XLSum English31.6%SR

Math

GSM8k91.4%SR
GSM8K Chat81.9%SR

Reasoning

HellaSwag85.6%SR

AA Evaluation Indices

Math Index
11.0
Intelligence Index
7.6
Math 500
0.7
Mmlu Pro
0.7
Gpqa
0.5
Ifbench
0.3
Aime
0.2
Scicode
0.2
Tau2
0.2
Livecodebench
0.2
Aime 25
0.1
Lcr
0.1
Hle
0.0
Terminalbench Hard
0.0

LLM Stats Category Scores

Math
90
Language
80
Legal
70
Reasoning
70
Finance
70
Healthcare
70
General
50
Roleplay
10
Communication
10
Creativity
10

Pricing

Input Price$1.2 / 1M tokens
Output Price$1.2 / 1M tokens
Blended Price (3:1)$1.2 / 1M tokens

Speed

Tokens/sec295.6
Time to First Token4.91s
Time to Answer4.91s

Provider Price Ranking

Provider Price Ranking

2 providers

Cheapest: NanoGPTMost Expensive: NVIDIA
ProviderInputOutput
1NanoGPTCheapest
$0.357
$0.408
2NVIDIAPRIMARY
$1.2
$1.2

Compare pricing across different API providers for this model.

External Sources