LLMcompare

Search

Search for a command to run...

A · open · Kimi

Kimi K2.5

Moonshot

B · closed · Kimi

Kimi K2.6

Moonshot

3:10

overall capability wins (1 ties across 14 scored). Kimi K2.6 leads overall.

  • Reasoning1:6
  • Coding1:2
  • Tool use0:2
  • Pricing0:3
  • Speed0:1
  • Specs1:0

Verdict

Kimi K2.6 leads coding (widest gap: +20.2 on TermBench); Kimi K2.5 allows ~3.9x more max output (64K vs 16K); Kimi K2.5 ships open weights (Modified MIT).

  • Kimi K2.6 leads coding (widest gap: +20.2 on TermBench)
  • Kimi K2.5 allows ~3.9x more max output (64K vs 16K)
  • Kimi K2.5 ships open weights (Modified MIT)
  • Kimi K2.6 is ~1.2x cheaper on a blended token basis than Kimi K2.5

Cost for 1M in + 250K out

Illustrative chat workload at primary-provider list prices.

Kimi K2.5
$1.23
Kimi K2.6
$1

Kimi K2.6 is about 1.2x cheaper on this workload.

Price per 1M tokens

Full comparison

Organization

Moonshot

Moonshot

Family

Kimi

Kimi

License

Modified MIT

Proprietary

Open weights

Yes

No

Release

Jan 27, 2026

Apr 8, 2026

Knowledge cutoff

-

-

API / provider

Moonshot

Moonshot

Modalities

text, image → text

text, image → text

Specs

Context window

256K

256K

tie

Max output

64K

16K

A +48K

Parameters

1.0T (32B act.)

—

—

Pricing

Input $/1M

$0.60

$0.50

B +$0.10

Output $/1M

$2.50

$2

B +$0.50

Blended $/1M (3∶1)

$1.07

$0.88

B +$0.20

Speed

tok/s

70

70

tie

TTFT (s)

0.45

0.4

B +0.05

Reasoning

MMLU-Pro

87.1

—

—

GPQA Diamond

87.9

91.1

B +3.2 pts

Humanity's Last Exam

30.7

37.5

B +6.8 pts

AIME 2025

96.1

—

—

MATH-500

—

—

—

Humanity's Last Exam (with tools)

—

—

—

AA-Omniscience Accuracy

35.2

32.6

A +2.6 pts

AA-LCR v1.1

78.0

81.0

B +3.0 pts

CritPt

3.1

8.0

B +4.9 pts

MMMU-Pro

75.4

79.4

B +4.0 pts

IFBench

70.2

76.0

B +5.8 pts

Chartography

—

—

—

Chartography (With Tools)

—

—

—

Coding

SWE-bench Verified

76.8

—

—

SWE-bench Pro

50.7

—

—

SWE-bench Multilingual

67.3

—

—

LiveCodeBench

85.0

—

—

Terminal-Bench 2.1

45.7

65.9

B +20.2 pts

Aider Polyglot

—

—

—

Terminal-Bench 3

—

—

—

BigCodeBench

—

—

—

SciCode

49.0

51.5

B +2.5 pts

CursorBench

—

—

—

SWE-Rebench

49.0

46.5

A +2.5 pts

NL2Repo-Bench

—

—

—

DeepSWE

—

—

—

WebDev Arena

—

1,509

—

Terminal-Bench 4.0

—

—

—

CursorBench 4.0

—

—

—

FrontierCode v1.1 (Main)

—

—

—

Arena

LMArena Elo

—

1,460

—

Kimi K2.5

Context
256K / 64K out
Parameters
1.0T (32B act.)
Price
$0.60 / $2.50
Speed
70 tok/s · 0.45s TTFT
Modalities
text, image → text
License
Modified MIT

Prior Kimi open MoE generation - strong long-context agentic coding before K3.

Kimi K2.6

Context
256K / 16K out
Parameters
—
Price
$0.50 / $2
Speed
70 tok/s · 0.4s TTFT
Modalities
text, image → text
License
Proprietary

Post-K2.5 Moonshot API SKU with longer context and stronger multimodal agent tooling.

Benchmark charts

Winner bars are emphasized. Per-benchmark deltas sit above each chart.

Reasoning & knowledge

GPQA: B +3.2HLE: B +6.8CritPt: B +4.9AA-LCR: B +3.0Omniscience Acc.: A +2.6MMMU-Pro: B +4.0IFBench: B +5.8

Coding

TermBench: B +20.2SciCode: B +2.5SWE-Rebench: A +2.5

Tool use & function calling

Toolathlon V.: B +25.0τ³-Banking: B +9.1

Pick a different pair · Back to catalog