vs
0:14
overall capability wins (0 ties across 14 scored). Kimi K3 leads overall.
- Reasoning0:6
- Coding0:3
- Arena0:1
- Tool use0:2
- Composite indices0:2
- Pricing3:0
- Speed2:0
- Specs0:2
Verdict
Kimi K3 leads coding (widest gap: +165.5 on WebDev Elo); Kimi K3 ranks higher on LMArena (+24 Elo); Kimi K3 offers ~4.1x larger context (1.0M vs 256K).
- Kimi K3 leads coding (widest gap: +165.5 on WebDev Elo)
- Kimi K3 ranks higher on LMArena (+24 Elo)
- Kimi K3 offers ~4.1x larger context (1.0M vs 256K)
- Kimi K3 allows ~3.9x more max output (64K vs 16K)
- Kimi K3 ships open weights (Kimi K3 License)
- Kimi K2.6 is ~6.9x cheaper on a blended token basis than Kimi K3
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Kimi K2.6
- $1
- Kimi K3
- $6.75
Kimi K2.6 is about 6.8x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Moonshot
Moonshot
Family
Kimi
Kimi
License
Proprietary
Kimi K3 License
Open weights
No
Yes
Release
Apr 8, 2026
Jul 16, 2026
Knowledge cutoff
-
-
API / provider
Moonshot
Moonshot
Modalities
text, image → text
text, image, video → text
Specs
Context window
256K
1.0M
Max output
16K
64K
Parameters
—
2.8T (104B act.)
Pricing
Input $/1M
$0.50
$3
Output $/1M
$2
$15
Blended $/1M (3∶1)
$0.88
$6
Speed
tok/s
70
60
TTFT (s)
0.4
0.6
Reasoning
MMLU-Pro
—
—
GPQA Diamond
91.1
93.5
Humanity's Last Exam
37.5
46.9
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
32.6
47.6
AA-LCR v1.1
81.0
88.7
CritPt
8.0
23.4
MMMU-Pro
79.4
80.5
IFBench
76.0
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
65.9
85.0
Aider Polyglot
—
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
51.5
59.5
CursorBench
—
60.8
SWE-Rebench
46.5
—
NL2Repo-Bench
—
—
DeepSWE
—
69.0
WebDev Arena
1,509
1,674
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,460
1,485
| Metric | Kimi K2.6 | Kimi K3 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Moonshot | Moonshot | — |
| Family | Kimi | Kimi | — |
| License | Proprietary | Kimi K3 License | — |
| Open weights | No | Yes | — |
| Release | Apr 8, 2026 | Jul 16, 2026 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Moonshot | Moonshot | — |
| Modalities | text, image → text | text, image, video → text | — |
| Specs | |||
| Context window | 256K | 1.0M | B +793K |
| Max output | 16K | 64K | B +48K |
| Parameters | — | 2.8T (104B act.) | — |
| Pricing | |||
| Input $/1M | $0.50 | $3 | A +$2.50 |
| Output $/1M | $2 | $15 | A +$13 |
| Blended $/1M (3∶1) | $0.88 | $6 | A +$5.13 |
| Speed | |||
| tok/s | 70 | 60 | A +10 |
| TTFT (s) | 0.4 | 0.6 | A +0.20 |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | 91.1 | 93.5 | B +2.4 pts |
| Humanity's Last Exam | 37.5 | 46.9 | B +9.4 pts |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 32.6 | 47.6 | B +15.0 pts |
| AA-LCR v1.1 | 81.0 | 88.7 | B +7.7 pts |
| CritPt | 8.0 | 23.4 | B +15.4 pts |
| MMMU-Pro | 79.4 | 80.5 | B +1.1 pts |
| IFBench | 76.0 | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | 65.9 | 85.0 | B +19.1 pts |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | 51.5 | 59.5 | B +8.0 pts |
| CursorBench | — | 60.8 | — |
| SWE-Rebench | 46.5 | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | 69.0 | — |
| WebDev Arena | 1,509 | 1,674 | B +166 Elo |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,460 | 1,485 | B +24 Elo |
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.
Kimi K3
- Context
- 1.0M / 64K out
- Parameters
- 2.8T (104B act.)
- Price
- $3 / $15
- Speed
- 60 tok/s · 0.6s TTFT
- Modalities
- text, image, video → text
- License
- Kimi K3 License
Largest open-weight model (2.8T MoE / 104B active) — official card reports strong reasoning and agentic coding results.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.