Model library

Open weights, published rates.

Every model below runs on the Akash GPU marketplace and bills per token. Rates, context limits and the precision label are on the card, and on the model’s own page next to a request you can copy.

Available now

Live endpoints with published per-token rates.

Kimi K3.1 Pro

2.8T / 104B active · 1M context

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

Moonshot AIMultimodalfp4
Input / 1M$1.75
Output / 1M$10.00
Cache read / 1M$0.18
View model

GLM 5.3

753B · 1M context

Complex coding and long-horizon agent work.

Z.aiChatfp4
Input / 1M$0.19
Output / 1M$4.40
Cache read / 1M$0.19
View model

Qwen 3.8 27B

27B · 262K context

Dense vision-language model for coding and research.

AlibabaMultimodalfp8
Input / 1M$0.225
Output / 1M$1.98
Cache read / 1M$0.05
View model

Qwen 3.6 35B A3B

35B / 3B active · 262K context

Multimodal hybrid MoE for reasoning and code.

AlibabaMultimodalfp8
Input / 1M$0.10
Output / 1M$0.90
Cache read / 1M$0.05
View model

GPT OSS 120B

120B / 5.1B active · 128K context

Open-weight MoE for reasoning and tool use.

OpenAIChatfp4
Input / 1M$0.037
Output / 1M$0.187
Cache read / 1M$0.037
View model

GPT OSS 20B

21B / 3.6B active · 128K context

Small MoE for agentic tool use and code execution.

OpenAIChatfp4
Input / 1M$0.018
Output / 1M$0.09
Cache read / 1M$0.009
View model

Llama 3.3 70B

70B · 128K context

Instruct-tuned for general reasoning and chat.

MetaChatfp8
Input / 1M$0.20
Output / 1M$0.52
Cache read / 1M$0.10
View model

Coming soon

Announced on the model library, not yet serving. No rate is published until an endpoint is live.

ModelProviderContextRequest
MiniMax M2.7MiniMax—Request model
DeepSeek V4 Flash 0731DeepSeek1MRequest model
Kimi K3Moonshot AI1MRequest model
Nemotron 3.5 LightningNVIDIA262KRequest model
Kimi K2.7 CodeMoonshot AI256KRequest model
GLM 5.2Z.ai1MRequest model
Kimi K2.6Moonshot AI256KRequest model
Gemma 4 26B A4B ITGoogle256KRequest model
DeepSeek V4 FlashDeepSeek262KRequest model
Qwen 3.5 35B A3BAlibaba262KRequest model
MiniMax M2.5MiniMax204KRequest model
Qwen3 30B A3BAlibaba32KRequest model
DeepSeek V3.2DeepSeek128KRequest model
Gemma 4 31B ITGoogle256KRequest model

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

Side by side

Put two rates next to each other.

Add a closed first-party API to see the gap on the same axes — and the row neither of them can fill in.

Open models on AkashML

Closed first-party APIs

3 of 5 columns

AttributeKimi K3.1 ProAkashMLClaude Sonnet 5.5AnthropicGPT-6.1 SolOpenAI
WeightsOpenClosedClosed
Parameters2.8T / 104B active——
Context1M1M1.05M
Precision publishedfp4 · disclosedno published weights to labelno published weights to label
Input / 1M$1.75$2.00$2.00
Output / 1M$10.00$10.00$10.00
Cache read / 1M$0.18$0.10$0.10
Blended / 1M3:1 input to output$3.813$4.0001.0x$4.0001.0x
LinksModel page Vendor pricing Vendor pricing

First-party list rates verified Nov 13, 2026, each column links to its source

Blended is 0.75 × input + 0.25 × output — a mix for ranking, not a rate. Cache reads are excluded from it.