Models / Kimi K2 (0905)

Moonshot AI

Kimi K2 (0905)

Kimi K2 (0905) is strongest at writing; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is Qwen3.5-9B, at $0.11 per million tokens against $1.07.

Superseded. Kimi K3 is Moonshot AI's newer model in this line, at $6.00 blended: see Kimi K3 or compare the two.

Results 16 Nov 2026 · 19 results · 4 sources

Test Kimi K2 (0905) on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.60 in · $2.50 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
15.3
Artificial Analysis
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Moonshot AI, based in China
Weights published
Released
5 Sep 2025
262,144 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 111th of 185Claude Opus 5.5 – $1.07 blended, 57th percentilecheapest: Mistral Nemo –
Models priced $1–3 per million tokens 30th of 34Muse Spark 1.3 – sets this group –
Models with similar intelligence sets this group – $1.07 blended, 83rd percentilecheapest: Qwen3.5-9B –
Models from developers based in China 47th of 75MiMo-V2.6-Pro – $1.07 blended, 74th percentilecheapest: Qwen3.5-9B –
Open-weights models 48th of 104MiMo-V2.6-Pro – $1.07 blended, 79th percentilecheapest: Mistral Nemo –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $1–3 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls Not measured
  • tau2-bench: has not measured this model.
  • Berkeley Function Calling Leaderboard (BFCL) V4: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 125th of 187 available · 33rd of 35 at its price
BenchmarkKimi K2 (0905)Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 15.3 111th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 6.4% 141st of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 76.7% 105th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 74th of 169 available · 21st of 28 at its price
BenchmarkKimi K2 (0905)Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,419 80th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 54.4 46th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 128th of 139 available · 26th of 26 at its price
BenchmarkKimi K2 (0905)Rank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 17.9% 66th of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,403 96th of 109 Gemini 4 Argon 1,511
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Kimi K2 (0905) at $1.07 blended, Intelligence Index 15.3. "Similar intelligence" means within 4 points or better.

Where to run it

1 host on OpenRouter at the standard service tier, with a median of $1.07 blended per million tokens; quantisation where the host states it: FP8. Cheapest first.

HostQuantisationInputOutputBlended
NovitaFP8$0.60$2.50$1.07

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Kimi K2 (0905)Muse Spark 1.3
most intelligent at $1–3 per million tokens
Qwen3.5-9B
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
MiMo-V2.6-Pro
most intelligent from China
Price per million tokens $1.07 $2.00$0.11$8.00$0.54
Tokens a second – 2541109643
Intelligence Index 15.3 48.113.357.646.3
Spring Prompt overall – 59–82–
Rank among available models, by use case
Product listings – 17th–13th–
Decks from an analysis – 25th–7th–
User surveys – ––4th–
Marketing planning – 25th–9th–
Agents and tool use – 13th111th3rd23rd
Professional work – 8th–3rd–
Reasoning and knowledge 125th of 187 11th116th1st13th
Writing 74th of 169 7th112th5th23rd
Sticking to the facts 128th of 139 11th–3rd10th
Speed under pressure – 33rd–––

Shaded figures are better than Kimi K2 (0905)'s. A dash means no result.

Where it has been measured

30th most intelligent of the 34 models priced $1–3 per million tokens. Among models available today, in the bottom quarter for sticking to the facts.

Artificial Analysis lists it but has not published a speed for it yet

Every result

19 published results from 4 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,403
1,389–1,418
151 of 182 Kimi K2 (0905) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,404
1,390–1,417
153 of 406 Kimi K2 (0905) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,381
1,366–1,397
141 of 411 Kimi K2 (0905) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,417
1,393–1,441
158 of 363 Kimi K2 (0905) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,390
1,379–1,401
158 of 413 Kimi K2 (0905) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,419
1,412–1,425
131 of 413 Kimi K2 (0905) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,382
1,370–1,394
155 of 412 Kimi K2 (0905) 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 7 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 15.3 252 of 413 Kimi K2 (0905) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 76.7% 204 of 360 Kimi K2 (0905) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 6.4% 323 of 419 Kimi K2 (0905) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 41.7% 193 of 288 Kimi K2 (0905) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 53.0% 269 of 407 Kimi K2 (0905) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 23.5% 138 of 282 Kimi K2 (0905) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 73.4% 119 of 286 Kimi K2 (0905) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

UGI Leaderboard · reported by UGI Leaderboard · 3 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 22.0% 210 of 379 Kimi K2 (0905) 15 Oct 2025
UGI: style adherencescore from 0 to 1, higher is better 0.31 324 of 379 Kimi K2 (0905) 15 Oct 2025
UGI: writing scorescore out of 100, higher is better 54.4 135 of 379 Kimi K2 (0905) 15 Oct 2025

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 98.6% 76 of 108 Kimi K2 (0905) 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 17.9% 97 of 108 Kimi K2 (0905) 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Will Kimi K2 (0905) work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.