vs
5:1
overall capability wins (0 ties across 6 scored). Kimi K2 leads overall.
- Reasoning4:0
- Specs1:1
Verdict
Kimi K2 edges reasoning & knowledge; Kimi Linear 48B A3B Instruct offers ~8.2x larger context (1.0M vs 128K).
- Kimi K2 edges reasoning & knowledge
- Kimi Linear 48B A3B Instruct offers ~8.2x larger context (1.0M vs 128K)
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Kimi K2
- $1.23
- Kimi Linear 48B A3B Instruct
- -
Price per 1M tokens
Full comparison
Organization
Moonshot
Moonshot
Family
Kimi
Kimi
License
Modified MIT
MIT
Open weights
Yes
Yes
Release
Jul 11, 2025
Oct 30, 2025
Knowledge cutoff
-
-
API / provider
Moonshot
No primary API price listed
Modalities
text → text
text → text
Specs
Context window
128K
1.0M
Max output
8K
-
Parameters
1.0T (32B act.)
48B (3B act.)
Pricing
Input $/1M
$0.60
—
Output $/1M
$2.50
—
Blended $/1M (3∶1)
$1.07
-
Speed
tok/s
50
-
TTFT (s)
0.55
-
Reasoning
MMLU-Pro
81.1
—
GPQA Diamond
76.6
41.2
Humanity's Last Exam
7.4
2.5
AIME 2025
49.5
—
MATH-500
97.4
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
27.4
—
AA-LCR v1.1
53.0
28.0
CritPt
—
—
MMMU-Pro
—
—
IFBench
41.5
28.1
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
65.8
—
SWE-bench Pro
27.7
—
SWE-bench Multilingual
47.3
—
LiveCodeBench
53.7
—
Terminal-Bench 2.1
—
—
Aider Polyglot
59.1
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
34.5
—
CursorBench
—
—
SWE-Rebench
34.6
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
—
—
| Metric | Kimi K2 | Kimi Linear 48B A3B Instruct | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Moonshot | Moonshot | — |
| Family | Kimi | Kimi | — |
| License | Modified MIT | MIT | — |
| Open weights | Yes | Yes | — |
| Release | Jul 11, 2025 | Oct 30, 2025 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Moonshot | No primary API price listed | — |
| Modalities | text → text | text → text | — |
| Specs | |||
| Context window | 128K | 1.0M | B +921K |
| Max output | 8K | - | — |
| Parameters | 1.0T (32B act.) | 48B (3B act.) | A +952 |
| Pricing | |||
| Input $/1M | $0.60 | — | — |
| Output $/1M | $2.50 | — | — |
| Blended $/1M (3∶1) | $1.07 | - | — |
| Speed | |||
| tok/s | 50 | - | — |
| TTFT (s) | 0.55 | - | — |
| Reasoning | |||
| MMLU-Pro | 81.1 | — | — |
| GPQA Diamond | 76.6 | 41.2 | A +35.4 pts |
| Humanity's Last Exam | 7.4 | 2.5 | A +4.9 pts |
| AIME 2025 | 49.5 | — | — |
| MATH-500 | 97.4 | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 27.4 | — | — |
| AA-LCR v1.1 | 53.0 | 28.0 | A +25.0 pts |
| CritPt | — | — | — |
| MMMU-Pro | — | — | — |
| IFBench | 41.5 | 28.1 | A +13.4 pts |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | 65.8 | — | — |
| SWE-bench Pro | 27.7 | — | — |
| SWE-bench Multilingual | 47.3 | — | — |
| LiveCodeBench | 53.7 | — | — |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | 59.1 | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | 34.5 | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | 34.6 | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | — | — | — |
Kimi K2
- Context
- 128K / 8K out
- Parameters
- 1.0T (32B act.)
- Price
- $0.60 / $2.50
- Speed
- 50 tok/s · 0.55s TTFT
- Modalities
- text → text
- License
- Modified MIT
Moonshot's trillion-parameter MoE (32B active) — open agentic coding before K2.5 / K3.
Kimi Linear 48B A3B Instruct
- Context
- 1.0M
- Parameters
- 48B (3B act.)
- Price
- —
- Speed
- —
- Modalities
- text → text
- License
- MIT
Moonshot AI's hybrid linear-attention MoE (48B total / 3B active) pairing Kimi Delta Attention with sparse global attention for cheaper long-context (up to 1M token) inference; no hosted API yet.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.