AiMatch

Hashference

Hashference routes requests to 61 AI models through an OpenAI- and Anthropic-compatible API. Developers and agent operators can fund requests with crypto, mining credits, or x402 per-request work.

Visit Hashference Compare alternatives Own this tool? Claim it
Hashference screenshot

Who it's for

  • Developers connecting apps that already use the OpenAI or Anthropic API to a catalog of models.
  • Agent operators using tools such as Claude Code, Cursor, Codex, OpenCode, or OpenClaw and looking for alternative ways to pay for model requests.
  • Miners with CPUs, GPUs, or ASICs who want accepted shares credited toward model inference.

How it fits your workflow

Get an API key, then fund the workspace with crypto, mining credits, or a staking plan. API keys in a workspace spend from the same balance.

Point an SDK or supported tool at Hashference and send requests through its OpenAI- or Anthropic-compatible API. For single-request payment, an app or agent can use x402 proof of work; miners can earn credits using the algorithms listed for their hardware.

Pricing

Qwen3.8 27B

$0.0018 per 1M input tokens; $0.022 per 1M output tokens

  • · Model pricing shown on the page

DeepSeek V4.1 Flash

$0.027 per 1M input tokens; $0.30 per 1M output tokens

  • · Model pricing shown on the page

GLM 5.3

$2.10 per 1M input tokens; $6.60 per 1M output tokens

  • · Model pricing shown on the page

Payment options

Mining credits, crypto top-ups, or per-request x402 payment; the page does not state a fixed price for these options

  • · Mine credits with supported CPU, GPU, or ASIC algorithms
  • · Buy credits with QUAI, Qi, stablecoins such as USDT and USDC, or other crypto
  • · Pay for an individual request using x402

Prices checked on Nov 5, 2026 from the vendor's site. They can change; confirm before you buy.

Key features

OpenAI- and Anthropic-compatible API
Connect applications that speak the OpenAI or Anthropic API to Hashference's model catalog.
Three payment methods
Pay by mining credits, buying credits with crypto, or having an app or agent pay for a single request using x402 proof of work.
Mining for credits
Point a CPU, GPU, or ASIC at Hashference. The page lists RandomX for CPUs, KawPoW for GPUs, and SHA-256 and Scrypt for ASICs; accepted shares become credits usable for any model.
Shared workspace balance
Every API key in a workspace spends from the same balance.
Model catalog and published token rates
The page lists 61 models and shows input and output prices per million tokens for featured models.

Works with

  • OpenAI API-compatible applications
  • Anthropic API-compatible applications
  • Developer tools and agents
  • Workflow and AI frameworks

Alternatives to consider

  • Token Harbor

    Choose it if you want a multi-provider model API that emphasizes model integrity, data privacy, transparent infrastructure, and responsive support rather than Hashference's mining and x402 payment options.

  • DDS Hub AI API Gateway

    Choose it if you specifically need access to Anthropic Claude and Codex GPT models with enterprise static IPs or Alipay and WeChat Pay, rather than Hashference's broader listed model catalog and mining-based credits.

Compare Hashference side by side

Getting started

  1. Get an API key.
  2. Fund it with crypto, mining credits, or a staking plan.
  3. Configure your SDK or supported tool to use Hashference, following the Quickstart.