impulse

Grok 4.20 Reasoning

Frontier lab
API id grok-4.20-reasoning
Azure AI Foundry retires it on 6 Apr 2027 (in 139 days). 5 hosts keep serving it.
Input / 1M
$1.25
at xAI
Output / 1M
$2.50
at xAI
Context window
1M
tokens
Max output
1M
tokens per reply
Hosts
6
9 listings
Released
-
date not published

Pricing

List prices at xAI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$1.25$2.50
Cached input
84% off input
$0.20
Batch
20% off
$1.00$2.00
Prompts over 200K tokens$2.50$5.00

What it costs a month

At xAI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$468
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$438
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$768
  • Overnight batch20,000 requests a day, 4K in / 0.4K out, batch$2,880
Use your own numbers

Where to run it

6 hosts serve Grok 4.20 Reasoning. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
xAILab API
grok-4.20-reasoning
$1.25 / $2.50
grok-4-20-reasoning
$1.25 / $2.50
Retires 6 Apr 2027
grok-4.20-reasoning
$1.25 / $2.50
xai.grok-4.20-reasoning
$1.25 / $2.50
PerplexityGateway
grok-4.20-reasoning
$1.25 / $2.50
AIHubMixGateway
grok-4-20-reasoning
$2.00 / $6.00
+92% vs headline
Show 3 more listings: regions, snapshots and variants
xAI
Alias
grok-4.20-reasoning-latest
$1.25 / $2.50
0309
grok-4.20-0309-reasoning
$1.25 / $2.50
Oracle OCI
0309
xai.grok-4.20-0309-reasoning
$1.25 / $2.50

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
TextImages
Produces
Text
Features
Prompt cachingBuilt-in web search

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than xAI.

Smaller xAI models

  • Grok Build 0.1
    xAI · on xAI
    $1.00 / $2.00
    Has every key capability
    20% less
  • Grok Code Fast
    xAI · on xAI
    $1.00 / $2.00
    Has every key capability
    20% less

Other frontier labs

  • Gemini 3.8 Flash
    Google · on Google AI Studio
    $0.75 / $3.75
    Has every key capability
    4% less
  • GPT-5.4 Nano
    OpenAI · on OpenAI
    $0.20 / $1.25
    Has every key capability
    70% less
  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    87% less
  • Gemini 3.5 Flash Lite
    Google · on Google AI Studio
    $0.30 / $2.50
    Has every key capability
    46% less

Open-weight picks

  • Kimi K2.7 Code
    Moonshot AI · on CLF AI Gateway
    $0.38 / $1.60
    Has every key capability
    56% less
  • Inkling Small
    Independent · on DeepInfra
    $0.45 / $1.20
    Has every key capability
    59% less
  • DeepSeek Flash
    DeepSeek · on DeepSeek
    $0.30 / $1.20
    Has every key capability
    66% less
  • Qwen3.7 Plus
    Alibaba Qwen · on Replicate
    $0.276 / $1.10
    Has every key capability
    69% less

Call it

Using xAI's API.

Python
import os
from openai import OpenAI

client = OpenAI(base_url="https://api.x.ai/v1", api_key=os.environ["XAI_API_KEY"])

completion = client.chat.completions.create(
    model="grok-4.20-reasoning",
    messages=[{"role": "user", "content": "Hello"}],
)
print(completion.choices[0].message.content)

Quick answers

How much does Grok 4.20 Reasoning cost?

On xAI, Grok 4.20 Reasoning costs $1.25 per million input tokens and $2.50 per million output tokens. Cached input costs $0.20 per million tokens.

Which APIs offer Grok 4.20 Reasoning?

Grok 4.20 Reasoning runs on 6 APIs: xAI, Azure AI Foundry, Vertex AI, Oracle OCI, Perplexity and AIHubMix.

What is Grok 4.20 Reasoning's context window?

1M tokens, with up to 1M tokens in one reply.

Is Grok 4.20 Reasoning being retired?

Azure AI Foundry retires it on 6 Apr 2027. 5 hosts keep serving it.

What are cheaper alternatives to Grok 4.20 Reasoning?

Models that cost less and keep most of what it can do: Grok Build 0.1 on xAI (20% less), Gemini 3.8 Flash on Google AI Studio (4% less), GPT-5.4 Nano on OpenAI (70% less) and Kimi K2.7 Code on CLF AI Gateway (56% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Grok 4.20 Reasoning's API model id?

Use grok-4.20-reasoning with xAI. Other hosts use their own ids: grok-4-20-reasoning on Azure AI Foundry, xai.grok-4.20-reasoning on Oracle OCI, grok-4-20-reasoning on AIHubMix.

Sources

Raw catalog entry for grok-4.20-reasoning on xAI
{
  "cache_read_input_token_cost": 2e-7,
  "cache_read_input_token_cost_batches": 1.6e-7,
  "input_cost_per_token": 0.00000125,
  "input_cost_per_token_batches": 0.000001,
  "litellm_provider": "xai",
  "max_input_tokens": 1000000,
  "max_output_tokens": 1000000,
  "mode": "chat",
  "output_cost_per_token": 0.0000025,
  "output_cost_per_token_batches": 0.000002,
  "source": "https://api.x.ai/v1/language-models",
  "supports_function_calling": true,
  "supports_reasoning": true,
  "supports_tool_choice": true,
  "supports_vision": true,
  "supports_web_search": true,
  "input_cost_per_token_above_200k_tokens": 0.0000025,
  "input_cost_per_token_above_200k_tokens_batches": 0.000002,
  "output_cost_per_token_above_200k_tokens": 0.000005,
  "output_cost_per_token_above_200k_tokens_batches": 0.000004,
  "cache_read_input_token_cost_above_200k_tokens": 4e-7,
  "cache_read_input_token_cost_above_200k_tokens_batches": 3.2e-7,
  "input_cost_per_image_token": 0.00000125,
  "supports_prompt_caching": true,
  "supports_response_schema": true
}