Agents and Inference.
Big models. Small bills.

GraphN runs open models behind one OpenAI-compatible API, with pre-built, customizable AI blueprints on top.

GraphN
import osfrom openai import OpenAIclient = OpenAI(    base_url="https://model.graphn.ai/v1",    api_key=os.environ["GRAPHN_API_KEY"],)reply = client.chat.completions.create(    model="deepseek-v4.1-flash",    messages=[{"role": "user", "content": "Hello"}],)

Build agents on the same models

Grow a model call into a workflow with tools and knowledge bases.

The workflow platform →
  • Support Router

    Route customer messages to the right specialist for orders, returns, or product questions

  • Company Knowledge Search

    Ask natural-language questions and get cited answers from your internal docs

  • Automated PR Review Bot

    Clone a repo, analyze a PR with a code-specialized model, synthesize review comments, and post them

graphn.ai
GraphN/Personal/Development WorkspaceD
DeepSeek V4.1 Flash Is Now Available on GraphNRead the announcement

Home

Requests19,52419,433 before · unc…
Org spend$67.15All workspaces in th…
Error rate4.5%884 errors · up 0.22…
TTFT p50320 msp95 463 ms · uncha…
Cache hit27%8.4M cached of 31.…
Tokens37.1M31.6M in · 5.6M out
per hour
SatSunMonTueWedThuFri
custom:cm-mock-1gpt-oss-120bimported:im-mock-1qwen3-80bqwen3.8-27b
220165110550
Click a bar, or drag across several, to open those requests in Logs
By modelUsage
ModelRequestsShareChange
qwen3.8-27b3,95820%+1%
gpt-oss-120b3,94920%+1%
RequestsLive
ModelKeyTTFT
11:36:05qwen3-80bCI pipeline312 ms
11:35:48qwen3-80bCI pipeline298 ms
11:35:31custom:cm-mo…Staging integ…341 ms
graphn.ai · Inference
DeepSeek V4.1 Flash1M context
Kimi K2.5256K context
Mistral Large 3256K context
Qwen3 VL 235B256K context
Qwen3.8 27B256K context
+ Your model$ graphn model deploy
Managed inference · OpenAI-compatible API
terminal
$ graphn model chat deepseek-v4.1-flash -m "Summarize this ticket"
The customer was charged twice at renewal. Refund one charge and confirm by email.
$ graphn model custom deploy --name support-8b \
    --hf-model-id Qwen/Qwen3-8B
✓ Custom model support-8b deployment started
Managed inference · OpenAI-compatible API

The built-in catalog

Chat & vision · 10

ModelContextThinkingToolsJSONVisionAudioInput · 1MOutput · 1M
DeepSeek V4.1 Flashdeepseek-v4.1-flash1M··$0.30$1.20
Kimi K2.5 (image)kimi-k2.5256K···$0.60$3.00
Mistral Large 3mistral-large-3256K····$0.50$1.50
Qwen3 VL 235B Instruct (image)qwen3-vl-235b256K···$0.53$2.66
Qwen3.8 27B (image+video)qwen3.8-27b256K·$0.45$3.20
gpt-oss 120Bgpt-oss-120b128K····$0.15$0.60
Qwen3 80B Instruct (131K context)qwen3-80b128K···$0.20$0.60
Qwen3.5 9B Vision (image+video)qwen3-vl256K·$0.10$0.15
Gemma 4 E4B (text+image+video+audio)gemma-4-e4b64K·$0.15$0.60
GLM-5.3glm-5.3250K···$1.40$4.40
Show all models
  1. 01

    Create an account

    Free to sign up.

  2. 02

    Add a card

    Get $5 in credits to start.

  3. 03

    Create an API key

    Point any OpenAI SDK at GraphN and pick a model by name.

Make your first call.

Free to sign up. $5 in credits to start.
Point any OpenAI SDK at GraphN.

Interested in building with Lightning?

Tell us what you need and we’ll reach out.

  • See a demo on workloads like yours
  • Get GPU pricing and capacity
  • Review security and compliance

Trusted by 450,000+ devs and AI teams

NVIDIACursorCiscoRunwaySupercellstability.aiBlack Forest Labs

You can also email us at sales@lightning.ai