Nousresearch

Nous: Hermes 3 70B Instruct

by Nousresearchtext->text1 endpoint
Neura Intelligence Index
42.2/ 100
Rank
#215
Confidence
high
Benchmarks
12(31% coverage)

Overview

Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the...

Capabilities

Text generation
Long context
Top-P sampling
Stop sequences
Deterministic seed

Modalities

Input
Text
Output
Text

Technical Specifications

Context Window
131.1K tokens
Max Output
16.4K tokens
Knowledge Cutoff
2023-12-31
Tokenizer
Llama3
Uptime (24h)
10000.0%
Parameters
70B
Blended Cost
$0.88/1M tokens
License
llama-3.1Open Weights
Country
🇺🇸US
Model Family
Llama 3.1
Release Date
July 23, 20242y

Supported Parameters

frequency_penaltylogit_biasmax_tokensmin_ppresence_penaltyrepetition_penaltyseedstopstructured_outputstemperaturetop_ktop_p

Benchmark Performance

general

BenchmarkScoreSource
Chatbot Arena Elo
1207
aggregated
IFEval
83.6
aggregated
MMLU-Pro
66.4
aggregated

coding

BenchmarkScoreSource
HumanEval
80.5
aggregated

math

BenchmarkScoreSource
GSM8K
95.1
aggregated
MATH
68
aggregated

reasoning

BenchmarkScoreSource
ARC-Challenge
94.8
aggregated
BigBench-Hard
81.2
aggregated
GPQA Diamond
46.7
aggregated

language

BenchmarkScoreSource
HellaSwag
94.4
aggregated
WinoGrande
85.3
aggregated

safety

BenchmarkScoreSource
TruthfulQA
60.1
aggregated

Performance Over Time

Pricing

Per 1M tokens
Input$0.70
Output$0.70
Added
August 18, 2024
Last synced 9/20/2026