Grok 4.20 Multi Agent Experimental beta 0304
xAI responsesGrok 4.20 Multi Agent Experimental beta 0304 is a xAI responses-API model.It supports a 1,000,000-token context windowwith up to 1,000,000 output tokens.Input is priced at $1.25/M tokens and output at $2.50/M tokens. Capabilities include vision, reasoning, prompt caching. Route Grok 4.20 Multi Agent Experimental beta 0304 via Future AGI's Agent Command Center for unified observability, caching, and 15 routing strategies including cost-optimized fallback.
Estimate Grok 4.20 Multi Agent Experimental beta 0304 spend
Pick a workload, fine-tune the sliders, and see the monthly bill.
Estimate uses $1.25/M input · $2.50/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.
Pricing
Per-token rates, expressed in USD per 1M tokens. Verified Oct 22, 2026.
| Input | $1.25/M | |
| Output | $2.50/M | |
| Blended (3:1 in:out) | $1.56/M Cross-model comparable price | |
| Cached input | $0.200/M |
Context & limits
- Context window
- 1,000,000 tokens
- Max input
- 1,000,000 tokens
- Max output
- 1,000,000 tokens
- Modalities
- vision, text
Capabilities
- Function calling — not advertised
- Parallel tool calls — not advertised
- Vision input ✓ supported
- Audio input — not advertised
- Audio output — not advertised
- PDF input — not advertised
- Streaming ✓ supported
- Structured output ✓ supported
- Prompt caching ✓ supported
- Reasoning ✓ supported
Strengths & caveats — when to pick Grok 4.20 Multi Agent Experimental beta 0304
Data-driven from Grok 4.20 Multi Agent Experimental beta 0304's percentile within responses peers.
Where it's strong
- +long-context tasks — context window in the top 8% of peers
- +long-form generation — 1,000,000-token max output, top-0% of peers
Watch out for
- No major caveats flagged from public spec.
Benchmarks pending
We haven't logged public benchmark scores for Grok 4.20 Multi Agent Experimental beta 0304 yet. Have one to contribute? Submit a source — citations help us prioritise.
Call Grok 4.20 Multi Agent Experimental beta 0304 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 Multi Agent Experimental beta 0304 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-multi-agent-experimental-beta-0304",
messages=[{"role": "user", "content": "Hello, Grok 4.20 Multi Agent Experimental beta 0304!"}],
)
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-multi-agent-experimental-beta-0304",
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 ↗Explore similar models
Grok 4.20 Multi Agent Experimental beta 0304 doesn't have a public Arena ELO score yet, so we group by provider only — quality-tier comparisons need a benchmark. Side-by-side links appear where a comparison page exists; otherwise, open the model profile.
- Grok 4.20 Multi Agent beta 0309xAI · $1.25/M in · $2.50/M out · 1,000,000 ctxView model →
- Grok 4.20 Multi AgentxAI · $1.25/M in · $2.50/M out · 1,000,000 ctxView model →
- Grok 4.20 Multi Agent 0309xAI · $1.25/M in · $2.50/M out · 1,000,000 ctxView model →
- Grok 4.20 Multi Agent beta latestxAI · $1.25/M in · $2.50/M out · 1,000,000 ctxView model →
FAQ
How much does Grok 4.20 Multi Agent Experimental beta 0304 cost?
Input is priced at $1.25 per 1M tokens and output at $2.50 per 1M tokens (xAI, last verified Oct 22, 2026).
What is the context window of Grok 4.20 Multi Agent Experimental beta 0304?
Grok 4.20 Multi Agent Experimental beta 0304 supports a 1,000,000-token context window with up to 1,000,000 output tokens.
Does Grok 4.20 Multi Agent Experimental beta 0304 support function calling?
Grok 4.20 Multi Agent Experimental beta 0304 does not currently advertise function-calling support. For agentic workloads, prefer a tool-calling-capable model and route via Agent Command Center for fallback.
Is Grok 4.20 Multi Agent Experimental beta 0304 good for production?
Grok 4.20 Multi Agent Experimental beta 0304 is well-suited for long-context tasks — context window in the top 8% of peers and long-form generation — 1,000,000-token max output, top-0% of peers.
How can I route to Grok 4.20 Multi Agent Experimental beta 0304 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 Multi Agent Experimental beta 0304
Official sources, independent benchmarks, and pricing aggregators — no random search-engine guesses.
Third-party evals — verify the marketing.
Cross-check our number against the rest of the ecosystem.