Grok 4.20 beta 0309 Reasoning
xAI chatGrok 4.20 beta 0309 Reasoning is a xAI chat model.It supports a 2,000,000-token context windowwith up to 2,000,000 output tokens.Input is priced at $2.00/M tokens and output at $6.00/M tokens. Capabilities include function calling, vision, reasoning, prompt caching. Route Grok 4.20 beta 0309 Reasoning via Future AGI's Agent Command Center for unified observability, caching, and 15 routing strategies including cost-optimized fallback.
Estimate Grok 4.20 beta 0309 Reasoning spend
Pick a workload, fine-tune the sliders, and see the monthly bill.
Estimate uses $2.00/M input · $6.00/M output. Provider pricing changes. Production costs vary with retries, streaming overhead, and tool-call rounds.
Want this for free? Cache + route via Agent Command Center — first 100K requests and 100K cache hits free every month.
Grok 4.3 (xAI) sits in the same Chatbot Arena tier (±80 ELO of Grok 4.20 beta 0309 Reasoning) and runs ~43% cheaper for a typical RAG workload — $723/mo vs $1,278/mo at 3K in / 400 out · 5K reqs/day.
Quality match is gated on real benchmarks; the CTA disappears when no comparable peer exists.
Pricing
Per-token rates, expressed in USD per 1M tokens. Verified May 12, 2026.
| Input | $2.00/M | |
| Output | $6.00/M | |
| Cached input | $0.200/M |
Limits
- Context window
- 2,000,000 tokens
- Max input
- 2,000,000 tokens
- Max output
- 2,000,000 tokens
- Modalities
- vision, text
Capabilities
- Function calling ✓ supported
- Parallel tool calls — not advertised
- Vision input ✓ supported
- Audio input — not advertised
- Audio output — not advertised
- PDF input — not advertised
- Streaming ✓ supported
- Structured output — not advertised
- Prompt caching ✓ supported
- Reasoning ✓ supported
Where it's strong
- +long-context tasks — context window in the top 1% of peers
- +long-form generation — 2,000,000-token max output, top-0% of peers
- +prompt caching — only 23% of chat models on Future AGI advertise this
Watch out for
- !strict structured output — no JSON-schema enforcement, expect retry loops
Benchmark scores
Reported public benchmark numbers. Each row links to the source.
Call Grok 4.20 beta 0309 Reasoning via Agent Command Center
One OpenAI-compatible endpoint. Routing, fallback, semantic caching, guardrails, and cost tracking come along for the ride. First 100K requests + 100K cache hits free every month.
agentcc / @agentcc/client). Per-call metadata — provider, cost, latency, cache hit, request id — is returned on x-agentcc-* response headers, so any HTTP client can read it.# Grok 4.20 beta 0309 Reasoning via the Agent Command Center Python SDK
# pip install agentcc
import os
from agentcc import AgentCC
client = AgentCC(
api_key=os.environ["AGENTCC_API_KEY"], # from app.futureagi.com → Settings → API Keys
base_url="https://gateway.futureagi.com/v1",
)
resp = client.chat.completions.create(
model="xai/grok-4-20-beta-0309-reasoning",
messages=[{"role": "user", "content": "Hello, Grok 4.20 beta 0309 Reasoning!"}],
)
print(resp.choices[0].message.content)
print(f"Tokens: {resp.usage.total_tokens}")
# Per-call gateway metadata is returned on x-agentcc-* response headers.
# When you need it programmatically, use .with_raw_response to get them:
raw = client.chat.completions.with_raw_response.create(
model="xai/grok-4-20-beta-0309-reasoning",
messages=[{"role": "user", "content": "Same call, but I want the headers."}],
)
print("Provider:", raw.headers.get("x-agentcc-provider"))
print("Latency:", raw.headers.get("x-agentcc-latency-ms"), "ms")
print("Cost: ", raw.headers.get("x-agentcc-cost"), "USD")
print("Cache: ", raw.headers.get("x-agentcc-cache"))AGENTCC_API_KEY with a key fromapp.futureagi.com.Gateway docs ↗Advanced: fallback + cache config (YAML) ▸
strategy: cost-optimized
targets:
- model: grok-4-20-beta-0309-reasoning
provider: xai
weight: 80
fallbacks:
- model: grok-4-3
provider: xai
- model: claude-opus-4-6-20260205
provider: anthropic
guardrails: [pii, prompt-injection, secrets]
cache: { exact: true, semantic: true }Compare with similar models
Grouped by Chatbot Arena tier (Grok 4.20 beta 0309 Reasoning sits at 1477 ELO).
Arena ELO within −25, ≤70% of price
- Claude Opus 4.6Azure AI Foundry · $5.00/M in · $25.00/M out · 200,000 ctx
- Claude Opus 4.6 (2026-02-05)Anthropic · $5.00/M in · $25.00/M out · 1,000,000 ctx
- Gemini 3.1 Pro previewGoogle Vertex AI · $2.00/M in · $12.00/M out · 1,048,576 ctx
- Claude Opus 4.7Azure AI Foundry · $5.00/M in · $25.00/M out · 200,000 ctx
25–100 ELO lower, ≤50% of price
FAQ
How much does Grok 4.20 beta 0309 Reasoning cost?
Input is priced at $2.00 per 1M tokens and output at $6.00 per 1M tokens (xAI, last verified May 12, 2026).
What is the context window of Grok 4.20 beta 0309 Reasoning?
Grok 4.20 beta 0309 Reasoning supports a 2,000,000-token context window with up to 2,000,000 output tokens.
Does Grok 4.20 beta 0309 Reasoning support function calling?
Yes — Grok 4.20 beta 0309 Reasoning supports function (tool) calling.
Is Grok 4.20 beta 0309 Reasoning good for production?
Grok 4.20 beta 0309 Reasoning is well-suited for long-context tasks — context window in the top 1% of peers and long-form generation — 2,000,000-token max output, top-0% of peers. Consider alternatives if you need strict structured output — no JSON-schema enforcement, expect retry loops.
How can I route to Grok 4.20 beta 0309 Reasoning with fallback?
Use Agent Command Center: a single OpenAI-compatible endpoint that supports cost-optimized routing, latency-aware retries, model fallback, and shadow traffic. Configure once, swap models without app changes.
Useful links for Grok 4.20 beta 0309 Reasoning
Official sources, independent benchmarks, and pricing aggregators — no random search-engine guesses.
Start here — vendor docs are the only authoritative spec.
Third-party evals — verify the marketing.
Cross-check our number against the rest of the ecosystem.