OpenAI & Anthropic drop-in replacement

Inference at thespeed of thought.

Serve frontier open-weight models through OpenAI-compatible endpoints. Quantization labels published externally for every one, and pricing set by a market rather than a margin.

OpenAI SDK compatible • Quantization published externally • Prompt caching included

Start chatting with your prompt.

Describe what you need — then run it for real. Same OpenAI-compatible endpoint, frontier open models, up to 85% below first-party API prices.

What each model is for

Different work,
different models,
one price sheet.

Each model below is built for a different kind of work: agentic engineering, tool use, deep reasoning, multimodal code, everyday chat and more. Pick by fit, then check the price underneath.

Kimi K3.1 Pro

2.8T / 104B active · 1M context

Coding, 3D creation and long-horizon agent work.

Input / 1M$1.75
Output / 1M$10.00
Cache read / 1M$0.18
Try in Playground

Qwen 3.8 27B

27B parameters · 262K context

Dense vision-language model for coding and research.

Input / 1M$0.225
Output / 1M$1.98
Cache read / 1M$0.05
Try in Playground

Qwen 3.6 35B A3B

35B / 3B active · 262K context

Multimodal hybrid MoE for reasoning and code.

Input / 1M$0.10
Output / 1M$0.90
Cache read / 1M$0.05
Try in Playground

GPT OSS 120B

120B / 5.1B active · 128K context

Open-weight MoE for reasoning and tool use.

Input / 1M$0.037
Output / 1M$0.187
Cache read / 1M$0.037
Try in Playground

Llama 3.3 70B

70B parameters · 128K context

Instruct-tuned for general reasoning and chat.

Input / 1M$0.20
Output / 1M$0.52
Cache read / 1M$0.10
Try in Playground

Don’t see the model you need?

Every model page takes integration requests. The library grows from what people are actually building on.

Request Model Integration
Point your tools hereOpenAI + Anthropic compatible
// ~/.claude/settings.json
{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.akashml.com/anthropic",    "ANTHROPIC_AUTH_TOKEN": "akml-…",    "ANTHROPIC_DEFAULT_SONNET_MODEL": "zai-org/GLM-5.3"
  }
}
// opus and haiku map the same way
client = OpenAI(
    base_url="https://api.akashml.com/v1",    api_key=os.environ["AKASHML_API_KEY"],
)
r = client.chat.completions.create(
    model="Qwen/Qwen3.8-27B",
    messages=[{"role": "user", "content": "…"}],
)
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.akashml.com/v1",  apiKey: process.env.AKASHML_API_KEY,
});
curl https://api.akashml.com/v1/chat/completions \
  -H "Authorization: Bearer $AKASHML_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"Qwen/Qwen3.8-27B",
      "messages":[{"role":"user","content":"…"}]}'
Existing SDKs, existing code. Only the endpoint, the key and the model id change.

Zero-migration coding agents

Keep Claude Code.
Change one file.

Same agent workflow, same terminal commands. Map Anthropic model tiers onto open weights like Kimi K3.1 Pro and GLM 5.3.

Preset workloads
Priced on Qwen 3.8 27B
3

Monthly token assumptions

Input · 120M
$27
Output · 24M
$48
Cache reads · 1.4B
$68

AkashML cost

$143/mo

First-party API

$615/mo

Claude Sonnet 5.5 list: $2 / $10 / $0.10 cache

Net difference

$472/mo77% lower

Get Your API Key
Coding: Anthropic's Claude Code average, $150–250 a developer a month. List prices; outputs differ by model.

Spend estimator

See what your agent stack actually costs.

Agentic coding loops re-read the same context on every turn, which is where the bill actually comes from. Estimate it against published list rates before you commit to anything.

Pricing, with the label

Every price cell carries a precision label.

Model · $/1M tokensAkashMLTogetherNebiusAmazon Bedrock
GLM 5.3

$0.19 in

$4.40 out

$0.19 cache read

fp4
$1.40 in$4.40 outunknown$1.40 in$4.40 outfp4$1.68 in$5.28 outunknown
GPT OSS 120B

$0.037 in

$0.187 out

$0.037 cache read

fp4
$0.15 in$0.60 outunknown$0.15 in$0.60 outfp4$0.15 in$0.60 outunknown
Kimi K3.1 Pro

$1.75 in

$10.00 out

$0.18 cache read

fp4
$2.00 in$10.00 outunknown$2.00 in$10.00 outfp4not served

Rates read Nov 16, 2026 · competitor cells link to OpenRouter

$0.05 / 1M

Per 1M cached input tokens on Qwen 3.8 27B — about 78% below fresh input.

1M token context

Full context on Kimi K3.1 Pro and GLM 5.3 — whole repos in one prompt.

Proof on independent compute

Razer Powers Viral AI
Inference on Akash

To scale its global AVA Mini campaign, Razer integrated its open-source AIKit platform with the Akash independent compute network. By pooling distributed high-performance consumer GPUs behind a single managed endpoint, the engineering team achieved reliable elastic scaling with zero manual infrastructure intervention.

Read Case Study

$0.01

per generated image

3.24s

avg. end-to-end response time

3–15x

lower inference costs than centralized APIs

Pricing and precision

The price, and where to check it.

A rate means nothing until you know what’s behind it. Quietly narrowed weights cost output quality that never shows up on an invoice — which is why OpenRouter publishes a quantization label for every endpoint we run, independently of us.

ModelInput / 1MOutput / 1MCache read / 1MContextAction
Kimi K3.1 Pro$1.75$10.00$0.181MTest
GLM 5.3$0.19$4.40$0.191MTest
Qwen 3.8 27B$0.225$1.98$0.05262KTest
Qwen 3.6 35B A3B$0.10$0.90$0.05262KTest
GPT OSS 120B$0.037$0.187$0.037128KTest
GPT OSS 20B$0.018$0.09$0.009128KTest
Llama 3.3 70B$0.20$0.52$0.10128KTest
See the full pricing page

Per-model rates, the estimator, and how we compare against first-party APIs.

“The future of AI isn’t just better models – it’s efficient infrastructure. With Razer AIKit, many use cases already run locally. With Akash, we extend that into a decentralised cloud to scale efficiently.”

QUYEN QUACH

Vice President of Software, Razer

What you’re actually buying

Every other inference provider is a store. AkashML is a market.

A store sets its price from corporate margins and capital overhead. A market discovers what compute actually costs by letting hardware operators compete for your workload — and publishes the result somewhere you can check it without taking our word for it.

Multi-provider reverse auction

Providers compete for your workload

Capacity comes from independent operators bidding against each other for the job, so no single vendor sets the price.

Marketplace pricing

Priced by a market, not a margin

Rows carrying a time-limited rate are labelled with list struck through — when one ends you pay the list rate shown beside it, nothing more.

Independent verification

Published where we can’t edit it

Our prices, context limits and quantization labels sit on OpenRouter’s provider page, next to every other host of the same weights. The network’s own numbers come from Akash’s public dashboard API. Neither runs on our infrastructure.

Compare us on OpenRouter

Drop-in compatibility

Zero SDK rewrites

Keep your codebase, your Claude Code setup, your Cursor agents and your OpenAI SDK. Swap the base URL and you’re running at up to 85% below first-party API prices.

Direct developer channel

A shared channel, not a ticket form

Support runs in the community channel. Ask in the open and get an answer there — no portal, no case number, no form you can’t see the queue behind. Join the channel ↗

Scope and limits

Easier to read this now than to find out in week three. We’d rather be straight about the boundaries than win a build you migrate away from in a quarter.

What AkashML is.

Managed inference, per token

No GPU to provision and no capacity to reserve up front.

OpenAI-compatible

Change base_url and api_key. The rest of your code is untouched.

No lock-in

That same base URL points anywhere else. Open weights, standard endpoints, nothing proprietary to unpick.

Works with the tools you already use

Claude Code, Cursor, Cline, OpenHands, any standard SDK.

Prompt caching where supported

Cache reads run about 78% below fresh input — $0.05 against $0.225.

Quantization you can check

OpenRouter publishes a label for every endpoint we run, independently of us.

Direct developer support

Ask in the shared community channel instead of filing a ticket.

What AkashML isn’t.

Not a place for your own weights

We run a curated library, not your container.

Not SOC 2 or HIPAA certified yet

Need a signed certificate today? We aren't your host.

Not a padded catalogue

No dead models inflating a count. The library grows on request.

Not a free tier that scales

The playground is for evaluating. Production runs on a paid account.

Not SLA-backed today

Uptime is published on OpenRouter, so you can judge it before committing.

FAQs

If something here isn’t covered, the docs have the detail — or ask in the shared community channel and get an answer from someone who works on it.

What is AkashML?+

AkashML is managed inference for open-weight models, running on GPUs from the Akash Network marketplace. You call the models through an OpenAI-compatible API or an Anthropic-compatible Messages endpoint and pay per token.

Which models can I use?+

Every model we currently serve is listed in the model library, with its rates on the pricing page. The library is curated rather than exhaustive. If the model you need isn't there, request it; we review requests regularly when deciding what to add.

Do I need to change my code to use AkashML?+

Very little. If you use an OpenAI SDK, point its base URL at https://api.akashml.com/v1 and swap in your AkashML API key and a model id. Claude Code and other Anthropic clients use https://api.akashml.com/anthropic instead; see the Claude Code setup.

Where are datacenters located?+

Inference runs on GPUs supplied by independent providers on the Akash Network. The network's current providers and their regions are published on Akash's network stats.

Where can I find documentation?+

Full documentation, guides, and API references are available in our docs.

How is pricing calculated?+

Pricing is per 1M tokens, with separate rates for input and output, and a cache-read rate where the model supports prompt caching. Each model's rates are on the pricing page. You only pay for what you use, with no minimum commitment.

Are there free credits available?+

Yes. New accounts start on a free tier with free credit and no card needed. It covers selected models in the Playground, with per-request limits. Adding credits unlocks every model and API keys. If you need more to evaluate a larger workload, contact us.

Can I predict my monthly costs?+

Yes. Multiply your expected input, output and cache-read tokens by the model's published rates. The spend estimator on the pricing page does this for common workloads.

What payment methods are accepted?+

We accept card payments. Credits can be topped up manually or automatically when your balance runs low.

Start building.

Create a free account, copy your API key, and point your SDK at one new base URL. Nothing else in your code has to change.

Get Your API KeyContact sales
The AkashML playground