Baltor Get started

Endpoints

Charm Hyper

A hosted service with its own key and prices. This page lists the addresses it answers, the facts its documentation states, and the setup of each harness.

Addresses and facts

  • OpenAI Chat Completions: https://hyper.charm.land/v1 models.dev, read
Authentication
Authorization: Bearer, with the key in HYPER_API_KEY models.dev, read
Structured output
Unknown
Tool calling
23 of 23 listed models The share of this provider's models that models.dev records with tool calling. models.dev, read
Rate limits
Unknown
Prices
Unknown
Data retention
Unknown
Documentation
Read the page models.dev, read
Output limits Baltor recorded
Unknown

Harness setup

Replace deepseek-v4-flash with the model you want.

OpenCode

Put this in opencode.json in your project folder:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "hyper": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Charm Hyper",
      "options": {
        "baseURL": "https://hyper.charm.land/v1",
        "apiKey": "{env:HYPER_API_KEY}"
      },
      "models": {
        "deepseek-v4-flash": {
          "name": "deepseek-v4-flash"
        }
      }
    }
  }
}
  • OpenCode reads any OpenAI-compatible address through the @ai-sdk/openai-compatible package, and an address that speaks the Responses API through @ai-sdk/openai.

From OpenCode documentation, read .

Pi

Put this in ~/.pi/agent/models.json:

{
  "providers": {
    "hyper": {
      "baseUrl": "https://hyper.charm.land/v1",
      "api": "openai-completions",
      "apiKey": "$HYPER_API_KEY",
      "models": [
        {
          "id": "deepseek-v4-flash"
        }
      ]
    }
  }
}
  • The apiKey field can name an environment variable as $NAME.

From Pi documentation, read .

Codex

Codex speaks only the Responses API, and Charm Hyper documents no Responses address. A gateway that offers one can sit in between.

  • Codex speaks the Responses API only: responses is the one supported wire API of a custom provider. Ollama and LM Studio are built in and start with --oss.

From Codex documentation, read .

Claude Code

Claude Code sends Anthropic Messages requests, and Charm Hyper documents no such address. A gateway that translates to that API can sit in between.

  • Claude Code sends Anthropic Messages requests to ANTHROPIC_BASE_URL. Anthropic says it does not support routing Claude Code to models other than Claude through any gateway, so some features may not work with another model.

From Claude Code documentation, read .

Models it lists

Prices in US dollars per million tokens, input and output, as models.dev, read records them.

ModelInputOutputContextAs of
DeepSeek V4 Flash deepseek-v4-flash0.200 USD0.400 USD1,000,000 older than 30 days
DeepSeek V4 Flash 0731 deepseek-v4-flash-07310.440 USD1.32 USD1,000,000 older than 30 days
DeepSeek V4 Pro deepseek-v4-pro2.40 USD4.80 USD1,000,000 older than 30 days
DeepSeek V4 Pro 0813 deepseek-v4-pro-08131.44 USD4.31 USD1,000,000 older than 30 days
DeepSeek V4.1 Flash deepseek-v4.1-flash0.300 USD1.20 USD1,048,576
Gemma 4 26B A4B IT gemma-4-26b-a4b-it0.098 USD0.334 USD256,000 older than 30 days
GLM-5.2 glm-5.21.52 USD4.79 USD1,000,000 older than 30 days
GLM-5.3 glm-5.31.52 USD4.79 USD1,000,000
GLM-5.3-Flash glm-5.3-flash0.163 USD0.544 USD1,048,576
GPT OSS 120B gpt-oss-120b0.178 USD0.680 USD131,072 older than 30 days
Inkling inkling1.09 USD4.41 USD1,048,576
Kimi K2 Thinking kimi-k2-thinking0.600 USD2.50 USD262,144
Kimi K2.7 Code kimi-k2.7-code1.03 USD4.36 USD256,000 older than 30 days
Kimi K3 kimi-k33.27 USD16.33 USD1,048,576 older than 30 days
MiniMax-M2.7 minimax-m2.70.484 USD1.85 USD262,100 older than 30 days
MiniMax-M3 minimax-m30.327 USD1.31 USD512,000 older than 30 days
Qwen3.7 Flash qwen3.7-flash0.200 USD0.800 USD1,000,000 older than 30 days
Qwen3.7 Max qwen3.7-max2.50 USD7.50 USD1,000,000 older than 30 days
Qwen3.7 Plus qwen3.7-plus1.20 USD4.80 USD1,000,000 older than 30 days
Qwen3.8 2.4T A95B qwen3.8-2.4t-a95b2.00 USD6.00 USD1,000,000
Qwen3.8 27B qwen3.8-27b0.500 USD3.00 USD1,000,000
Qwen3.8 Flash qwen3.8-flash0.150 USD0.470 USD1,000,000
Qwen3.8 Max Preview qwen3.8-max2.00 USD6.00 USD1,000,000 older than 30 days

Sources of this page