Is Thalam really OpenAI-compatible?

Yes. Change base_url and api_key only. LangChain, LlamaIndex, and every OpenAI SDK work immediately.

How is Thalam priced?

Per-token pricing at the rate quoted for each model. No platform fee, no subscription, no seats. Credits never expire and you only pay for what you use.

Do you store prompts?

No. Prompts pass through in real time and are not logged or stored. Only metadata (model, token count, cost, timestamp) is recorded for billing.

How do I migrate from the OpenAI SDK?

Change the base_url to https://api.thalam.ai/v1 and swap your API key. Every OpenAI SDK works unchanged — LangChain, LlamaIndex, the official Python and Node libraries.

Every leading AI model. One API. One bill.

Pay per token. No minimums. No lock-in.

Get your free API key Browse all models

OpenAI

Anthropic

Google

xAI

Start here

Browse all models

Nano Banana Pro

NEWImage

Nano Banana Pro — Google's Gemini 3 Pro Image, frontier text-to-image with best-in-class in-image text.

Context

—

Max output

—

Latency

Slow

Input price

$0.134 / image

Output price

—

Test in Playground

Claude Opus 4.8

NEWTextArabic

Anthropic's newest frontier model. Highest capability for complex reasoning, long-form, and agentic tasks.

Context

Max output

32K

Latency

Medium

Input price

$5.00 / 1M

Output price

$25.00 / 1M

Test in Playground

Seedance 2.0

NEWVideo

ByteDance's Seedance 2.0 — multimodal video generation (text/image/reference → video) with optional synchronized audio.

Context

—

Max output

15s

Latency

Slow

Input price

$0.84 / video

Output price

—

Test in Playground

Why Thalam

Freedom, without the sprawl.

One bill, every model.

DeepSeek, Qwen, Kimi, Kling, Kolors — one key, one dashboard, one invoice at the end of the month.

thalam.Apr 2026

DeepSeek V3.2$12.40

Qwen3.5$04.10

Kimi K2.5$08.20

Kling V3$03.00

Kolors$00.80

Total$28.50

Governance built in.

Per-key spend caps, real-time alerts, usage analytics. Kill a runaway key without redeploying.

key_prod_8e3f

Active

Spend cap$142 / $200

▸alert at $180

2 lines to switch.

Change base_url and key. Every OpenAI SDK works unchanged — LangChain, LlamaIndex, Instructor.

-base_url="api.openai.com/v1"

+base_url="api.thalam.ai/v1"

-api_key=os.environ["OPENAI_KEY"]

+api_key=os.environ["THALAM_KEY"]

Priced per token.

No seats, no minimums, no expiry. Credits work the same across every model.

$ / 1M tokensin · out

deepseek-v3.2$0.27$0.40

qwen3.5-122b$0.40$3.20

kimi-k2.5$0.62$2.47

Get started

From zero to your first request in a minute.

No demos to book, no contracts to sign, no waitlist to join.

Create your account

One click with Google, GitHub, or email.

Top up your balance

Start with $5 or $500. One balance that works everywhere — text, image, and video models.

Apr 18

$50

Apr 09

$20

Ship with your key

Drop your key into any OpenAI SDK and change the base URL. No new library to learn.

THALAM_API_KEY

••••••••••••••••

ENTERPRISE

Enterprise plans for teams in production.

Custom volume pricing

AED or USD invoicing on contractual terms

Higher rate limits

Formal SLA on request

Dedicated support channel

DEVELOPER FIRST

Integration in Seconds.

We are perfectly compatible with OpenAI client libraries like LangChain, Ollama, and standard REST clients. No learning curve.

Standardized JSON Payloads

Streaming & async support

Unified Billing & Metering

quickstart.py

# No new library — just point your existing OpenAI client here

from openai import OpenAI

client = OpenAI(

api_key="tl-live-...",

base_url="https://api.thalam.ai/v1"

)

response = client.chat.completions.create(

model="deepseek/deepseek-v3.2",

messages=[{"role": "user", "content": "Summarise this contract in Arabic"}]

)

# Same SDK, same call shape — point it at any model in the catalog.

print(response.choices[0].message.content)