MMODELYST
Models/Llama 3.1 Nemotron Instruct 70B
MDL

Llama 3.1 Nemotron Instruct 70B

NVIDIA

Verified against Artificial Analysis · Jul 13, 2026
Capability
40
percentile index
Price /M
$1.2 / $1.2
in / out
Context
Avg score
28.2
11 benchmarks
AA Intelligence
7.6
index
Speed
310 tok/s
TTFT 3.9s
Released
Oct 15, 2024
About

Llama 3.1 Nemotron Instruct 70B is a open-weights model in the Llama family from NVIDIA. Benchmarked on 11 evals, averaging 28.2.

Benchmark scores · 11
avg 28.2 — higher is better
MATH
73.3
MMLU-Pro
69
GPQA Diamond
46.5
IFBench
30.7
SciCode
23.3
τ²-bench
23.1
LiveCodeBench
16.9
AIME
11
AA Long-Context Reasoning
7
Humanity's Last Exam
4.6
Terminal-Bench Hard
4.5
Details
PublisherNVIDIA
FamilyLlama
Parameters
License
WeightsOpen
ReleasedOct 15, 2024
Performance & price
Output speed310 tok/s
Latency (TTFT)3.9s
Price in / out$1.2 / $1.2 per 1M
Blended price$1.2 / 1M
Math index11.0

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers
Artificial Analysis$1.2/M