Kimi K2.6

Moonshot AI chat

Kimi K2.6 is a Moonshot AI chat model.It supports a 262,144-token context windowwith up to 262,144 output tokens.Input is priced at $0.950/M tokens and output at $4.00/M tokens. Capabilities include function calling, vision, reasoning. Route Kimi K2.6 via Future AGI's Agent Command Center for unified observability, caching, and 15 routing strategies including cost-optimized fallback.

Pricing source: litellm Last verified: May 12, 2026 View source ↗
Cost calculator

Estimate Kimi K2.6 spend

Pick a workload, fine-tune the sliders, and see the monthly bill.

~3K in / ~400 out · 5K req/day
3,000
0262,144
400
0200,000
5,000
01,000,000
cached @ $0.1600/M
Per request
$0.004450
in $0.002850 · out $0.001600
Per day
$22.25
5,000 requests
Per month
$677
152,188 requests

Estimate uses $0.9500/M input · $4.00/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.

Pricing

Per-token rates, expressed in USD per 1M tokens. Verified May 12, 2026.

Input $0.950/M
Output $4.00/M
Cached input $0.160/M

Limits

Context window
262,144 tokens
Max input
262,144 tokens
Max output
262,144 tokens
Modalities
vision, video, 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 — not advertised
  • Reasoning ✓ supported

Where it's strong

  • +long-form generation — 262,144-token max output, top-1% of peers

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.

AIMEmath· 2026
Captured May 12, 2026
Chatbot Arena ELOgeneral· overall
Captured May 12, 2026
MathVisionmultimodal· w/ python
Captured May 12, 2026
GPQA Diamondreasoning
Captured May 12, 2026
LiveCodeBenchcode· v6
Captured May 12, 2026
SWE-bench Verifiedagent
Captured May 12, 2026
MMMU-Promultimodal· w/ python
Captured May 12, 2026
SWE-benchagent· Pro
Captured May 12, 2026
Humanity's Last Examreasoning· w/ tools
Captured May 12, 2026
SciCodecode
Captured May 12, 2026
Try it

Call Kimi K2.6 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.

SDK
Native Future AGI client (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.
# Kimi K2.6 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="moonshot/kimi-k2-6",
    messages=[{"role": "user", "content": "Hello, Kimi K2.6!"}],
)

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="moonshot/kimi-k2-6",
    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"))
Set AGENTCC_API_KEY with a key fromapp.futureagi.com.Gateway docs ↗
Advanced: fallback + cache config (YAML)
strategy: cost-optimized
targets:
  - model: kimi-k2-6
    provider: moonshot
    weight: 80
fallbacks:
  - model: claude-opus-4-6
    provider: azure-ai-foundry
  - 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 (Kimi K2.6 sits at 1461 ELO).

FAQ

How much does Kimi K2.6 cost?

Input is priced at $0.950 per 1M tokens and output at $4.00 per 1M tokens (Moonshot AI, last verified May 12, 2026).

What is the context window of Kimi K2.6?

Kimi K2.6 supports a 262,144-token context window with up to 262,144 output tokens.

Does Kimi K2.6 support function calling?

Yes — Kimi K2.6 supports function (tool) calling.

Is Kimi K2.6 good for production?

Kimi K2.6 is well-suited for long-form generation — 262,144-token max output, top-1% of peers. Consider alternatives if you need strict structured output — no JSON-schema enforcement, expect retry loops.

How can I route to Kimi K2.6 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 Kimi K2.6

Official sources, independent benchmarks, and pricing aggregators — no random search-engine guesses.