ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
xAI
Proprietary API xAI · grok-4.20-0309-reasoning · Text+Image+PDF

Grok 4.20 (Reasoning) Official Specifications, Pricing & Benchmarks

Grok 4.20 (Reasoning) (grok-4.20-0309-reasoning) by xAI, released 2026-03-09, featuring 1,000,000 context window, 30,000 max output tokens, and Text+Image+PDF input.

📋 Grok 4.20 (Reasoning) Official API & System Card Specifications

Verified against official xAI developer documentation & pricing

API Model ID: grok-4.20-0309-reasoning

1. Identity & Capabilities

API Spec
API Model ID
grok-4.20-0309-reasoning
Release Date
2026-03-09
Weight License
Proprietary (Undisclosed)
Native Reasoning
✅ Supported
Function Calling & JSON Schema
✅ 原生支持
Batch API 异步批处理
✅ Supported (50% off)

2. Context, Output & Modalities

Limits
Max Context Window
1,000,000 tokens (1000K)
Max Output Tokens
30,000 tokens
Input Modalities
Text + Image + PDF
Output Modalities
Text
Direct China Access
🌐 Global / OpenRouter
Deployment Mode
Cloud API only (Proprietary)

3. Official Pricing & Cache

USD / 1M
Input / 1M Tokens
$ 1.25
Cached Input / 1M
$ 0.2 -84%
Output / 1M Tokens
$ 2.5
Tiered Pricing Notes

🎛️ Official Sampling & Reasoning Parameters

Recommended generation settings & API parameter constraints

Official Reasoning & Tool Config
temperature=0.7 top_p=0.95 max_tokens=30,000

Closest Self-Hostable Open-Weight Alternatives to Grok 4.20 (Reasoning)

Open-weight models with full HuggingFace weights for private GPU / Mac deployment

Compare with these →

Frequently Asked Questions

Is Grok 4.20 (Reasoning) open-source or proprietary?

Grok 4.20 (Reasoning) (grok-4.20-0309-reasoning) is a proprietary commercial model by xAI. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.

What are the context limit and API pricing of Grok 4.20 (Reasoning)?

Grok 4.20 (Reasoning) supports 1000K input context and 30,000 max output tokens, priced at $1.25/1M input ($0.2/1M cached) and $2.5/1M output.