LLM ReferenceLLM Reference

Mistral NeMo Instruct (2407)

Capabilities

VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution

Providers(7)

Compare all →
ProviderInput (per 1M)Output (per 1M)Type
NVIDIA NIMProvisioned
Microsoft Foundry$0.3$0.3Provisioned
DeepInfra$2$4Serverless
Fireworks AI$0.9$0.9Serverless
Arcee AI$0.15$0.45Serverless
Novita AI$0.08$0.24Serverless
Replicate API$0.45$0.45Serverless

Benchmark Scores(4)

BenchmarkScoreVersionSource
Google-Proof Q&A57.1diamondOpen LLM Leaderboard
HellaSwag91.810-shotOpen LLM Leaderboard
HumanEval81.1pass@1Open LLM Leaderboard
Massive Multitask Language Understanding81.55-shotOpen LLM Leaderboard

Rankings

Specifications

FamilyNeMo
Released2024-07-18
Parameters12B
Context128K
ArchitectureDecoder Only
Specializationgeneral
Trainingfinetuning

Created by

Enterprise AI solutions for trust and transparency.

Paris, France
Founded 2023
Website