Novita AI

Researched 1d agoInference PlatformTier 3

Novita AI

CodingRAGAgentsLong contextVisionClassificationJSON / Tool useapi

Novita AI offers 90 tracked models (85 with output token pricing). This catalog covers coding, rag, and agents; open any model detail page for benchmarks, batch tiers, and migration prompts. 13 retired models are listed below.

Covers 7 workload areas across 90 available models and 13 retired; last verified 2026-11-18.

Use it for

  • Teams comparing token and batch pricing across this provider's models
  • Operators routing coding, rag, and agents workloads through this API

Do not use it for

  • Final benchmark picks without opening the relevant model detail page

Tracked models

90

Models available through this provider

Priced output routes

85

Models with output token pricing tracked

Cheapest output

$0.020

PaddleOCR VL on this route

Batch-ready models

0

No batch pricing tracked

Latest model release

2026-11-16

3d since newest release

Freshness

2026-11-18

Researched 1d ago

fresh

Information

TypeInference Platform
TierTier 3
Models90
CompanyNovita AI

Novita AI offers a GPU-based inference API for image, video, and language model generation with a broad catalog of open-source models.

Where this host wins

  • Coding: 37 tracked models with SWE-bench / HumanEval-style scores.
  • RAG: 58 tracked models with ruler / needle retrieval benchmarks.
  • Agentic: 47 tracked models with BFCL, tau-bench, and SWE-bench tool-use coverage.
  • Long-context: 65 tracked models with context-token or InfiniteBench-class signal.

Getting started

Verify: quotas and regions in the linked vendor documentation.

Platform Overview

Novita AI offers a GPU-based inference API for image, video, and language model generation with a broad catalog of open-source models.

Available Models(90)

View all →

All models available as Serverless

ModelInput (per 1M)Output (per 1M)
$0.070$0.281
$1.413$4.24
$1.25$3.75
$0.6$3.6
$0.14$0.28
$1.60$3.20
$2$6
$0.248$1.485
$1.38$4.4
$0.13$0.4
View full catalog →

Retired models(13)

View all →
ModelInput (per 1M)Output (per 1M)
Ring-2.6-1TDeprecated
$0.3$2.5
Ling-2.6-1TDeprecated
$0.3$2.5
Ling-2.6-FlashDeprecated
$0.1$0.3
Kimi K2.6Deprecated
Scheduled retirement: December 31, 2026
$0.8$3.4
Retired November 18, 2026
$0.6$2.4
$0.98$3.95
Qwen3-MaxDeprecated
Retired October 10, 2026
$2.11$8.45
$0.3$0.3
Phi-4 MiniDeprecated
$0.05$0.15
$0.08$0.24
View full catalog →

Where else to run this