DeepSeek
Reasoning
DeepSeek V4 Pro
A high-capacity reasoning route for agentic workflows, coding tasks, and production chat traffic through Jatevo.
Input
$1.74
Output
$3.48
Speed
Fast
Jatevo model catalog
Serverless model cards for the routes behind Jatevo.
Browse every playground-ready and request-access model route with provider logos, pricing markers, deployment type, and capability tags in one place.
Browse Models
Find model routes by provider, capability, or deployment style.
10 models
Z.ai
Code
GLM 5.2
GLM 5.2 is Jatevo's dedicated Z.ai route for long-running software, agent, and tool-use sessions.
Input
$1.40
Output
$4.40
Context
1M
NVIDIA
NewReasoning
NVIDIA Nemotron 3 Ultra 550B A55B NVFP4
Nemotron 3 Ultra is a 550B hybrid MoE model from NVIDIA, optimized for demanding multi-agent AI and complex reasoning tasks.
Input
$0.60
Output
$3.60
Speed
59 Tok/s
Jatevo Inference
NewReasoning
Kimi K3
Kimi K3 provides maximum-effort reasoning and a 1,048,576-token context window through Jatevo API Master.
Input
$0.75
Output
$3.50
Context
1M
Jatevo Inference
NewCode
Kimi K2.7 Code
Kimi K2.7 Code runs on Jatevo Inference for long-context software work, agent execution, and production chat workloads.
Input
$0.75
Output
$3.50
Focus
Agentic
Alibaba Cloud
Chat
Qwen 3.7 Max
A Qwen Max route for text-only enterprise workflows, exposed through Jatevo with scoped key enforcement.
Input
$1.25
Output
$3.75
Speed
Fast
Cerebras
Chat
Cerebras GLM 4.7
GLM 4.7 served on the existing Cerebras route for low-latency prompt iteration and streaming chat sessions.
Input
Low
Output
Low
Speed
Ultra-fast
Cerebras
NewChat
Cerebras Gemma 4 31B
Gemma 4 31B runs through the existing Cerebras playground route for fast streaming chat and long completion tests.
Input
$1.00
Output
$1.50
Output cap
Fast
Spark
NewChat
Spark Gemma 4 26B A4B
Spark Gemma 4 26B A4B is exposed through API Master as a serverless Gemma route for lightweight chat and generation workloads.
Input
Metered
Output
Metered
Route
Fast
RTX PRO 6000
NewChat
RTX Qwen3.6 35B A3B NVFP4
Qwen3.6 35B A3B NVFP4 runs on Jatevo's RTX PRO 6000 route for high-memory chat and long-context inference through API Master.
Input
Metered
Output
Metered
Context
262K