viv-fast — API, Pricing & Context Window | Vivgrid

viv-fast is Vivgrid's SLM for fast, high-volume workloads. It accepts text, image input, keeps a 1M-token context window and 256K-token max output, supports function calling, and is priced at $0.13 per 1M input tokens, $0.05 cached, and $0.40 output.

It is available on both OpenAI surfaces — Chat Completions (/chat/completions) and the Responses API (/responses) — so coding CLIs that expect the Responses wire format, such as Codex and Pi, work against the same key.

Specifications

Pricing

Pricing in USD per 1M tokens, matching the provider's rates.

Quick start

Call viv-fast through Vivgrid's unified, OpenAI-compatible endpoint. Get an API key from the Vivgrid Console.

curl https://api.vivgrid.com/v1/chat/completions \
  -H "Authorization: Bearer $VIVGRID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "viv-fast",
    "messages": [
      { "role": "user", "content": "Say hello in English, Chinese and Spanish." }
    ],
    "stream": true
  }'

Use viv-fast in your coding agent

viv-fast works with the same Vivgrid key and https://api.vivgrid.com/v1 endpoint as every other model in the catalog. Grab your API key from the Vivgrid Console, then drop one of the configurations below into your agent of choice.

Codex

Install Codex:

npm i -g @openai/codex

Add Vivgrid as a provider and make viv-fast the default model in ~/.codex/config.toml. Codex talks to Vivgrid over the Responses API:

model = "viv-fast"
model_provider = "vivgrid"
preferred_auth_method = "apikey"
forced_login_method = "api"

[features]
streamable_shell = true
unified_exec = true
shell_snapshot = true

[model_providers.vivgrid]
name = "Vivgrid"
base_url = "https://api.vivgrid.com/v1"
wire_api = "responses"
experimental_bearer_token = "viv-xxxxxxxxxxxxx"

Then start Codex in your project:

codex

Already using another Vivgrid model in Codex? Switch for a single run with codex -m viv-fast. See the full Codex tutorial.

OpenCode

Install OpenCode:

npm i -g opencode-ai

Declare viv-fast under the Vivgrid provider and make it the default model in your global OpenCode config, ~/.config/opencode/opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "model": "vivgrid/viv-fast",
  "provider": {
    "vivgrid": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Vivgrid",
      "options": {
        "baseURL": "https://api.vivgrid.com/v1",
        "apiKey": "{env:VIVGRID_API_KEY}"
      },
      "models": {
        "viv-fast": {
          "name": "viv-fast",
          "tool_call": true,
          "attachment": true,
          "modalities": { "input": ["text", "image"], "output": ["text"] },
          "limit": { "context": 1000000, "output": 256000 }
        }
      }
    }
  }
}

Export your key and launch OpenCode:

export VIVGRID_API_KEY="viv-xxxxxxxxxxxxx"
opencode

Use /models to switch between viv-fast and other Vivgrid models mid-session. See the full OpenCode tutorial.

Pi

Install Pi:

npm install -g --ignore-scripts @earendil-works/pi-coding-agent

Add viv-fast to the Vivgrid provider in ~/.pi/agent/models.json:

{
  "providers": {
    "vivgrid": {
      "baseUrl": "https://api.vivgrid.com/v1",
      "apiKey": "viv-xxxxxxxxxxxxx",
      "api": "openai-completions",
      "models": [
        {
          "id": "viv-fast",
          "input": ["text", "image"],
          "contextWindow": 1000000,
          "maxTokens": 256000
        }
      ]
    }
  }
}

Make it the model Pi starts with in ~/.pi/agent/settings.json:

{
  "defaultProvider": "vivgrid",
  "defaultModel": "viv-fast"
}

Then verify and start Pi:

pi --list-models vivgrid
pi

Or run a one-shot prompt without entering the TUI:

pi -p "explain the auth flow in this repo" --model vivgrid/viv-fast

Pi can also reach viv-fast over the Responses API — set "api": "openai-responses" on the model entry. See the full Pi tutorial.

Ideal use cases

• High-volume agent loops — cheap input and cached-input pricing keeps long, tool-heavy sessions affordable.
• Multimodal understanding — reason over screenshots, diagrams, and other images in the same request as text.
• Long-context work — feed large codebases, transcripts, or document sets into the 1M-token window.
• Coding CLIs on the Responses API — drop-in for tools such as Codex and Pi.

Related models

• gpt-6-astra — frontier model from OpenAI
• claude-fable-5.1 — frontier model from Anthropic
• gemini-3.8-flash — fast multimodal model from Google
• deepseek-v4.1-flash — open-sourced fast, low-cost coding model