Overview of supported foundation models, context window limits, and token pricing.

Models & Pricing

All models are available through the unified /v1/chat/completions endpoint. Pricing is transparently billed per million tokens with no markup multipliers, seats, or monthly commitments.

Foundation Models Catalog

Model ID	Creator	Context Window	Input / 1M	Output / 1M	Best For
`deepseek-v4.1-flash` ⭐	DeepSeek	1,000,000	$0.23	$0.90	Next-gen frontier flash, ultra fast, code reasoning
`glm-5.3-flash`	Zhipu AI	1,000,000	$0.23	$0.75	Fast agent loops, auxiliary tasks, sub-cent queries
`gpt-6-luna`	OpenAI	1,100,000	$0.15	$0.75	Sub-cent long-context reasoning & conversational apps
`claude-haiku-5-5`	Anthropic	1,000,000	$0.15	$0.75	Ultra-efficient agentic tool-use & fast refactoring
`qwen3.8-flash`	Alibaba	1,000,000	$0.23	$0.71	High-throughput code synthesis & tool-calling
`minimax-m3.1-flash`	MiniMax	1,000,000	$0.23	$0.90	Multilingual code analysis & agentic tasks
`mimo-v2.6-flash`	Xiaomi	1,000,000	$0.21	$0.42	Ultra-long 1M context document summarization
`mimo-v2.6-pro`	Xiaomi	1,000,000	$0.65	$1.31	High-precision 1M context complex logic
`kimi-k2.7-code`	Moonshot AI	262,144	$1.43	$6.00	Multi-file refactoring, tool calling in agents
`grok-4.7`	xAI	500,000	$3.00	$9.00	Frontier creative synthesis & direct instructions
`kimi-k3`	Moonshot AI	1,048,576	$4.50	$22.50	Frontier reasoning & long-horizon agent planning
`qwen3.8-max`	Alibaba	262,144	$3.00	$9.00	High-capacity reasoning & heavy code architecture
`gemini-4-argon-contributor`	Google	1,000,000	$6.00	$30.00	Multi-modal frontier contributor model
`hy4-preview`	OpenCode	1,000,000	$1.25	$4.50	Agentic coding preview model
`deepseek-v4-pro`	DeepSeek	1,000,000	$1.98	$5.94	High-fidelity coding and math
`glm-5.3`	Zhipu AI	1,000,000	$2.10	$6.60	Complex reasoning & technical analysis
`longcat-2.0`	Meituan	262,144	$1.13	$4.43	Document processing & multilingual analysis

Dynamic Catalog Discovery

You can discover the live catalog programmatically without authentication:

curl https://api.interference.dev/v1/models

Or query in Python:

from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.interference.dev/v1",
    api_key=os.environ.get("INTERFERENCE_API_KEY", "public")
)

models = client.models.list()
for m in models:
    print(m.id)

Model Recommendations

Terminal & Coding Agents (OpenCode, Hermes, Aider):
- Default: kimi-k2.7-code or qwen3-coder for precise edits and syntax awareness.
- Fast helper: glm-5.3-flash for linting, file search, and background checks.
Deep Technical Reasoning:
- glm-5.3 or kimi-k3 with full chain-of-thought tokens streamed via delta.reasoning_content.
Cost Optimization:
- glm-5.3-flash delivers 1,000,000 tokens of input for just $0.23.

Models & Pricing

Models & Pricing

Foundation Models Catalog

Dynamic Catalog Discovery

Model Recommendations

On this page