The arena · AI model review

Kimi K3.1 Pro

by Moonshot AI

Moonshot's post-trained Kimi K3: near Opus 5.5 on Terminal-Bench 4.0, hallucinations cut by two thirds, a third cheaper at $2/$10

Arena score 3.5/5
Reasoning4.0
Coding4.5
Writing4.0
Speed1.5
Value3.5
Visit Kimi K3.1 Proagentic coding outside US clouds3D modeling, CAD and spatial reasoning tasksfinancial research and office documentsteams planning to self-host a 3T-class open model
Price

$10/1M out

Official Moonshot API list price for kimi-k3.1-pro: $2/1M input (cache miss), $0.20/1M on cache hit, $10/1M output, 1M-token context; kimi-k3.1-pro-highspeed at $4/$20 (cache hit $0.20). Batch API at 60% of standard price. Kimi K3 stays listed at $3/$15. Verified on platform.kimi.ai/docs/pricing/chat, 2026-11-03.

Provider

Moonshot AI

Context window

1M tokens

Input price

$2/1M in

Output price

$10/1M out

Modalities

text, vision

Open weights

No

What is Kimi K3.1 Pro?

Released 2026-10-23 by Beijing-based Moonshot AI as the first model of its K3.1 series, Kimi K3.1 Pro is a post-trained version of Kimi K3 with substantially more reinforcement-learning compute across hundreds of domains. Moonshot calls it its most capable 3T-class open-weight model, with a 1M-token context window and text and image input; the full weights are promised by 2026-11-04. API ID kimi-k3.1-pro at $2/1M input ($0.20 on cache hit) and $10/1M output, a third below Kimi K3's $3/$15; a Highspeed variant (kimi-k3.1-pro-highspeed) runs 2 to 3 times faster at $4/$20. Batch inference costs 60% of the standard price. Moonshot reports 65.7% on Terminal-Bench 4.0 (Claude Opus 5.5: 66.4%, GPT-6 Astra: 58.2%) and a claim-level hallucination rate down from 23.4% (K3) to 8.3% on its own adversarial Hallu Bench Hard. Artificial Analysis scores it 46 on the Intelligence Index v5.0 (#23 of 60). Available on Kimi.ai, Kimi Work, Kimi Code and the Kimi API.

Kimi K3.1 Pro pros & cons

Pros

  • 65.7% on Terminal-Bench 4.0 per Moonshot, just behind Claude Opus 5.5 (66.4%) and ahead of GPT-6 Astra (58.2%)
  • Far fewer hallucinations than K3 on Moonshot's adversarial Hallu Bench Hard: 8.3% of claims wrong with tools vs 23.4% for K3, better than Opus 5.5 (15.0%)
  • A third cheaper than Kimi K3: $2/$10 per 1M tokens (cache hit $0.20), batch inference at 60% of list price
  • Strong at 3D and spatial work: 92.7 on BenchCAD with tools, level with Claude Fable 5.1 (92.6), and 72.8 without tools, next to Opus 5.5 (73.0)
  • Finance and office gains over K3: Financial Research 70.4 vs 58.8, and a state-of-the-art 64.4 on OfficeQA Pro v2 per Moonshot
  • Open weights promised by 2026-11-04, after Kimi K3's were published on Hugging Face

Cons

  • Very slow: 30.6 output tokens per second on Artificial Analysis, #193 of 195; the Highspeed variant doubles the price to $4/$20
  • 46 on the Artificial Analysis Intelligence Index v5.0 (#23 of 60), well below Claude Sonnet 5.5 (54) and GPT-6.1 Sol (55) sold at the same $2/$10
  • Most headline gains come from Moonshot's internal benchmarks (Kimi Code Bench, Hallu Bench, finance suites) that cannot be checked independently
  • Weights and license not yet published at review time, and the API runs under Beijing jurisdiction, a compliance hurdle for many EU and enterprise deployments
  • Cyber data was removed from training for the open release, so offensive-security capability sits below frontier closed models (81.0 vs 90.5 for the unfiltered checkpoint on Moonshot's xdev-bench-hard)

The arena’s verdict

Kimi K3.1 Pro is a real upgrade over K3 at a lower price: near Opus 5.5 on Moonshot's Terminal-Bench 4.0 figure, standout 3D and CAD skills, and a hallucination rate cut by two thirds on Moonshot's own stress test. Take it for agentic coding, 3D and financial-research work, and especially if you plan to self-host once the weights land. Skip it for anything interactive (30 tokens per second is among the slowest on the market), and compare carefully before buying on price: at the same $2/$10, Claude Sonnet 5.5 and GPT-6.1 Sol score clearly higher on Artificial Analysis and run four to five times faster. K3 users should switch: K3.1 Pro is cheaper and better on every benchmark Moonshot published.

Thumbs up or thumbs down

Cast your verdict

Would you recommend Kimi K3.1 Pro, or warn the crowd away?

50%crowd score · 0

Top Kimi K3.1 Pro alternatives

All alternatives
1
Gemini 3.8 Flash

Google workhorse September 2026: 1M context, $0.75/$3.75 promo pricing, 305 tokens/s output speed

$3.75/1M out60%(1)
Visit
2
Gemini 3 Pro

Google's Nov 2025 frontier model: 1M context, first past 1500 Elo on LMArena, replaced by Gemini 3.1 Pro.

$12/1M out (prompts ≤200K)57%(3)
Visit
3
Grok 4.3

xAI's 1M-context frontier model at $1.25/$2.50 per 1M tokens, tuned for agentic tool use.

$2.50/1M out
Visit

Compare Kimi K3.1 Pro head-to-head

Kimi K3.1 Pro: frequently asked questions

How much does Kimi K3.1 Pro cost per 1M tokens?

Kimi K3.1 Pro costs $2/1M in per 1M input tokens and $10/1M out per 1M output tokens. Official Moonshot API list price for kimi-k3.1-pro: $2/1M input (cache miss), $0.20/1M on cache hit, $10/1M output, 1M-token context; kimi-k3.1-pro-highspeed at $4/$20 (cache hit $0.20). Batch API at 60% of standard price. Kimi K3 stays listed at $3/$15. Verified on platform.kimi.ai/docs/pricing/chat, 2026-11-03.