Skip to content
AI Atlas
Family

Kimi

All 17 models in the terminal →

first seen 11 Sept 2026

family_01M2A55TXQZ1M3SD7NPQVJQKY2

Models
17
First release
9 Apr 2025
Latest release
13 Jun 2026
Parameters
9.77B – 2.78T
Artifacts
4
Providers
3

Releases

Family releases by statusActiveDeprecatedMay 25Jun 25Jul 25Aug 25Sept 25Oct 25Nov 25Dec 25Jan 26Feb 26Mar 26Apr 26May 26Jun 26Kimi-VL-A3B-Instruct — 9 Apr 2025Kimi-VL-A3B-Thinking — 9 Apr 2025Kimi-Audio-7B — 25 Apr 2025Kimi-Audio-7B-Instruct — 25 Apr 2025Kimi-Dev-72B — 16 Jun 2025Kimi-VL-A3B-Thinking-2506 — 21 Jun 2025Kimi-K2-Base — 3 Jul 2025Kimi K2 0711 — 11 Jul 2025Kimi K2 0905 — 3 Sept 2025Kimi-Linear-48B-A3B-Base — 30 Oct 2025Kimi-Linear-48B-A3B-Instruct — 30 Oct 2025Kimi K2 Thinking — 4 Nov 2025cerebras/Kimi-Linear-REAP-35B-A3B-Instruct — 6 Nov 2025Kimi K2.5 — 1 Jan 2026Kimi K2.6 — 14 Apr 2026Kimi K2.7 Code — 11 Jun 2026Kimi K3 — 13 Jun 2026

17 dated releases · dot size = best benchmark rank of the member

Members 17

Family members
ModelParamsContextOpennessLicenceReleasedStatusBest rankCompare
Kimi K3Moonshot AI2.78T1.05MOpen weightsOther13 Jun 2026Active#5 LiveBench Reasoning
Kimi K2.7 CodeMoonshot AI1.03T262.1KOpen weightsOther11 Jun 2026Active#23 Terminal-Bench
Kimi K2.6Moonshot AI1.03T262.1KOpen weightsOther14 Apr 2026Deprecated#14 τ²-bench
Kimi K2.5Moonshot AI1.03T262.1KOpen weightsOther1 Jan 2026Deprecated#7 SWE-bench Multilingual
cerebras/Kimi-Linear-REAP-35B-A3B-InstructCerebras Systems35.1B3B active—Open weightsMIT6 Nov 2025Active—
Kimi K2 ThinkingMoonshot AI1.03T262.1KOpen weightsOther4 Nov 2025Deprecated#19 SWE-bench Verified
Kimi-Linear-48B-A3B-BaseMoonshot AI49.1B3B active—Open weightsMIT30 Oct 2025Active—
Kimi-Linear-48B-A3B-InstructMoonshot AI49.1B3B active1MOpen weightsMIT30 Oct 2025Active#157 Terminal-Bench
Kimi K2 0905Moonshot AI1.03T262.1KOpen weightsOther3 Sept 2025Deprecated#109 Terminal-Bench
Kimi K2 0711Moonshot AI1.03T131.1KOpen weightsOther11 Jul 2025Deprecated#21 Aider polyglot
Kimi-K2-BaseMoonshot AI1.03T—Open weightsOther3 Jul 2025Active—
Kimi-VL-A3B-Thinking-2506Moonshot AI16.4B3B active—Open weightsMIT21 Jun 2025Active—
Kimi-Dev-72BMoonshot AI72.7B—Open weightsMIT16 Jun 2025Active—
Kimi-Audio-7BMoonshot AI9.77B—Open weightsMIT25 Apr 2025Active—
Kimi-Audio-7B-InstructMoonshot AI9.77B—Open weightsMIT25 Apr 2025Active—
Kimi-VL-A3B-InstructMoonshot AI16.4B3B active—Open weightsMIT9 Apr 2025Active—
Kimi-VL-A3B-ThinkingMoonshot AI16.4B3B active—Open weightsMIT9 Apr 2025Active—

Lineage among members 3 relations

Open in Graph →

Benchmark progress

LiveBench Reasoningbest #5

Best rank of the family on LiveBench Reasoning#50#40#30#20#10#0Apr 26Apr 26May 26May 26May 26May 26May 26Jun 26

SciCodebest #5

Best rank of the family on SciCode#40#30#20#10#0Apr 26Apr 26May 26May 26May 26May 26May 26Jun 26

LiveBench Agentic Codingbest #8

Best rank of the family on LiveBench Agentic Coding#50#40#30#20#10#0Apr 26Apr 26May 26May 26May 26May 26May 26Jun 26

LiveBench Languagebest #8

Best rank of the family on LiveBench Language#50#40#30#20#10#0Apr 26Apr 26May 26May 26May 26May 26May 26Jun 26

Artificial Analysis Intelligence Indexbest #9

Best rank of the family on Artificial Analysis Intelligence Index#250#200#150#100#50#0Oct 25Jan 26Apr 26

GPQA Diamondbest #10

Best rank of the family on GPQA Diamond#200#150#100#50#0Oct 25Jan 26Apr 26

Best rank (primary comparability group, current leaderboard) reached by any member released up to each date — lower is better, axis inverted. Ranks are today's ranks, not the ranks at release time.

Aggregates are computed from the members' sourced claims; members without a release date or parameter count are left out of the range and timeline rather than guessed. Methodology →