llmrun

Kimi K2.6 vs Kimi K2.7 Code

Kimi K2.6 leads llmrun Score 120.9 to 118.1 (ranges overlap); neither fits 24 GB.

Model A · Moonshot AI

Kimi K2.6
120.9llmrun Score

1026.9B · 33.3B active · 256K context · Other · Apr 2026

VisionFunctions

Model B · Moonshot AI

Kimi K2.7 Code
118.1llmrun Score

1026.9B · 33.3B active · 256K context · Other · Jun 2026

VisionCodeFunctions

Which should you choose?

Choose Kimi K2.6 if you want…

  • Higher llmrun Score

    120.9 vs 118.1, ranges overlap

  • Ahead on the benchmarks

    11 of 14 shared boards

Choose Kimi K2.7 Code if you want…

  • Also does code

    Kimi K2.6 doesn't list it

On your hardware

Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.

8 GB

RTX 4060 · 8 GB tier

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

12 GB

RTX 3060 12GB · 12 GB tier

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

16 GB

RTX 5060 Ti 16GB · 16 GB tier

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

24 GBreference

RTX 4090 · 24 GB tier

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

32 GB

RTX 5090 · 32 GB tier

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

48 GB

RTX A6000 · 48 GB tier

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

Mac 32 GB

Mac Mini M4 · Macs

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

Mac 64 GB

Mac Studio M4 Max · Macs

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

Mac 128 GB

Mac Studio M4 Max · Macs

Kimi K2.6

Doesn't fit

smallest quant needs 283.8 GB

Kimi K2.7 Code

Doesn't fit

smallest quant needs 283.8 GB

Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).

llmrun Score

with its 90% range · bars share one scale

Overall

Kimi K2.6

120.9118–123

Kimi K2.7 Code

118.1115–121

Coding

Kimi K2.6

118.7111–126

Kimi K2.7 Code

114.8110–120

Agents

Kimi K2.6

121.9115–129

Kimi K2.7 Code

118.7112–125

Math

Kimi K2.6

122.1119–125

Kimi K2.7 Code

120.7116–125

Science

Kimi K2.6

121.5110–133

Kimi K2.7 Code

117.7107–128

Reasoning

Kimi K2.6

117.7111–125

Kimi K2.7 Code

115.4106–125

Kimi K2.6 · #12 of 100 open models · #10 of 96 that fit ≤640 GB

Kimi K2.7 Code · #14 of 100 open models · #12 of 96 that fit ≤640 GB

Benchmarks

14 shared boards: Kimi K2.6 leads 11, Kimi K2.7 Code leads 3 · Benchmarks imported Oct 2026

GPQA Diamond

Kimi K2.6

90.8%#8/98 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

87.9%#15/98 open

IndependentEpoch AI · Jun 2026

LiveBench Coding

Kimi K2.6

78.6#6/19 open

IndependentLiveBench · Jun 2026

Kimi K2.7 Code

74.0#11/19 open

IndependentLiveBench · Jun 2026

LiveBench Reasoning

Kimi K2.6

79.4#12/19 open

IndependentLiveBench · Jun 2026

Kimi K2.7 Code

82.8#9/19 open

IndependentLiveBench · Jun 2026

AIME 2024/2025

Kimi K2.6

96.1%#6/86 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

95.6%#7/86 open

IndependentEpoch AI · Jun 2026

FrontierMath

Kimi K2.6

57.2%#8/17 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

54.0%#10/17 open

IndependentEpoch AI · Jun 2026

LiveBench Math

Kimi K2.6

84.3#14/19 open

IndependentLiveBench · Jun 2026

Kimi K2.7 Code

79.6#18/19 open

IndependentLiveBench · Jun 2026

SciCode

Kimi K2.6

53.5%#4/66 open

Lab-reportedEpoch AI · Apr 2026

Kimi K2.7 Code

47.8%#16/66 open

Lab-reportedEpoch AI · Jun 2026

SimpleQA

Kimi K2.6

34.9%#10/17 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

36.5%#9/17 open

IndependentEpoch AI · Jun 2026

FrontierMath Tier 4

Kimi K2.6

25.6%#7/13 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

12.2%#11/13 open

IndependentEpoch AI · Jun 2026

WeirdML

Kimi K2.6

55.9%#7/46 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

54.1%#8/46 open

IndependentEpoch AI · Jun 2026

18 more boards (other shared and single-model boards)
ALE-Bench

Kimi K2.6

1092.7#7/43 open

Lab-reportedEpoch AI · Apr 2026

Kimi K2.7 Code

886.2#16/43 open

Lab-reportedEpoch AI · Jun 2026

Chess Puzzles

Kimi K2.6

20.0%#10/14 open

IndependentEpoch AI · Apr 2026

Kimi K2.7 Code

14.0%#13/14 open

IndependentEpoch AI · Jun 2026

CritPt

Kimi K2.6

8.0%#13/66 open

Lab-reportedEpoch AI · Apr 2026

Kimi K2.7 Code

10.0%#11/66 open

Lab-reportedEpoch AI · Jun 2026

Vending-Bench 2

Kimi K2.6

6204.6#3/17 open

Lab-reportedEpoch AI · Apr 2026

Kimi K2.7 Code

5082.9#6/17 open

Lab-reportedEpoch AI · Jun 2026

LMArena WebDev

Kimi K2.6

1509.6#14/52 open

ArenaLMArena · Oct 2026

Kimi K2.7 Code

1473.8#19/52 open

ArenaLMArena · Oct 2026

Only Kimi K2.7 Code (5)

VRAM by quantization

Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.

Q2_K

Kimi K2.6

440 GB358.2 GB file

32K 439 GB · 128K 446 GB

Kimi K2.7 Code

440 GB~436.4 GB file

32K 439 GB · 128K 446 GB

Q3_K_S

Kimi K2.6

453 GB444.4 GB file

32K 452 GB · 128K 459 GB

Kimi K2.7 Code

not listed

Q3_K_M

Kimi K2.6

504 GB466.9 GB file

32K 503 GB · 128K 510 GB

Kimi K2.7 Code

504 GB463.6 GB file

32K 503 GB · 128K 510 GB

Q4_0

Kimi K2.6

517 GB583.7 GB file

32K 516 GB · 128K 523 GB

Kimi K2.7 Code

not listed

Q4_K_M

Kimi K2.6

620 GB~616.1 GB file

32K 619 GB · 128K 626 GB

Kimi K2.7 Code

620 GB~616.1 GB file

32K 619 GB · 128K 626 GB

Q5_K_M

Kimi K2.6

736 GB~731.6 GB file

32K 734 GB · 128K 741 GB

Kimi K2.7 Code

736 GB~731.6 GB file

32K 734 GB · 128K 741 GB

Q6_K

Kimi K2.6

851 GB~847.2 GB file

32K 850 GB · 128K 857 GB

Kimi K2.7 Code

851 GB~847.2 GB file

32K 850 GB · 128K 857 GB

Q8_0

Kimi K2.6

1031 GB~1026.9 GB file

32K 1029 GB · 128K 1036 GB

Kimi K2.7 Code

1031 GB~1026.9 GB file

32K 1029 GB · 128K 1036 GB

11 more quants

IQ2_XXS

Kimi K2.6

286 GB271.5 GB file

32K 285 GB · 128K 292 GB

Kimi K2.7 Code

286 GB317.8 GB file

32K 285 GB · 128K 292 GB

IQ2_XS

Kimi K2.6

312 GB303.2 GB file

32K 311 GB · 128K 318 GB

Kimi K2.7 Code

not listed

IQ2_S

Kimi K2.6

325 GB308.8 GB file

32K 324 GB · 128K 330 GB

Kimi K2.7 Code

325 GB334.8 GB file

32K 324 GB · 128K 330 GB

IQ2_M

Kimi K2.6

350 GB341.7 GB file

32K 349 GB · 128K 356 GB

Kimi K2.7 Code

350 GB318.0 GB file

32K 349 GB · 128K 356 GB

IQ3_XXS

Kimi K2.6

402 GB426.6 GB file

32K 401 GB · 128K 407 GB

Kimi K2.7 Code

402 GB376.9 GB file

32K 401 GB · 128K 407 GB

IQ3_XS

Kimi K2.6

427 GB466.4 GB file

32K 426 GB · 128K 433 GB

Kimi K2.7 Code

not listed

IQ3_S

Kimi K2.6

440 GB405.3 GB file

32K 439 GB · 128K 446 GB

Kimi K2.7 Code

440 GB418.8 GB file

32K 439 GB · 128K 446 GB

IQ3_M

Kimi K2.6

466 GB488.1 GB file

32K 465 GB · 128K 472 GB

Kimi K2.7 Code

not listed

Q3_K_L

Kimi K2.6

530 GB487.5 GB file

32K 529 GB · 128K 536 GB

Kimi K2.7 Code

530 GB493.9 GB file

32K 529 GB · 128K 536 GB

IQ4_XS

Kimi K2.6

556 GB546.2 GB file

32K 555 GB · 128K 561 GB

Kimi K2.7 Code

556 GB495.1 GB file

32K 555 GB · 128K 561 GB

BF16

Kimi K2.6

2058 GB2053.2 GB file

32K 2056 GB · 128K 2063 GB

Kimi K2.7 Code

2058 GB~2053.8 GB file

32K 2056 GB · 128K 2063 GB

How to run each

Specifications

Specifications of Kimi K2.6 and Kimi K2.7 Code
SpecKimi K2.6Kimi K2.7 Code
PublisherMoonshot AIMoonshot AI
Parameters1026.9B1026.9B
Active parameters33B per token33B per token
ArchitectureKimiK25ForConditionalGenerationKimiK25ForConditionalGeneration
Context256K tokens256K tokens
ModalitiesVision, Function-callingVision, Code, Function-calling
LicenseOtherOther
ReleasedApr 2026Jun 2026

Frequently Asked Questions

Which is better, Kimi K2.6 or Kimi K2.7 Code?

Kimi K2.6 leads llmrun Score 120.9 to 118.1 (ranges overlap); neither fits 24 GB. Choose Kimi K2.6 for higher llmrun Score and ahead on the benchmarks; choose Kimi K2.7 Code for also does code.

Which needs less VRAM, Kimi K2.6 or Kimi K2.7 Code?

At Q4_K_M, Kimi K2.6 requires 620.0 GB while Kimi K2.7 Code needs 620.0 GB, making Kimi K2.6 the easier model to run on your local setup.

Can I run Kimi K2.6 or Kimi K2.7 Code on a 16 GB GPU?

Kimi K2.6 does not fit (smallest quant needs 283.8 GB); Kimi K2.7 Code does not fit (smallest quant needs 283.8 GB).

Which has a longer context window, Kimi K2.6 or Kimi K2.7 Code?

Kimi K2.6 supports 262,144 tokens and Kimi K2.7 Code supports 262,144 tokens.

Kimi K2.6 vs Kimi K2.7 Code — VRAM & GPU Comparison | llmrun