Kimi K2 Thinking: Price, Context, Benchmarks, and Release Details
52.1
SI Score (method si-v3-retained-evidence-2)
19% confidence 19 percent, Low confidence — 2 of 8 expected sources in
Coverage 15% of expected source weight · 100% confidence at 80% coverage
Pillar breakdown
Coding (weight 40 percent) 66.7
Math (weight 15 percent) —
Preference (weight 15 percent) —
Reasoning (weight 30 percent) —
Pillar weights: reasoning 30% · math 15% · coding 40% · preference 15%. Benchmark results use fixed 0–100 scales before averaging; incomplete evidence is shrunk toward 50.
Facts
- Input price / 1M
- not yet reported
- Output price / 1M
- not yet reported
- Context window
- 262Kmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Max output
- 262Kmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Released
- Nov 6, 2025models.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Open weights
- YesHugging Face HubPublic Hub repo with weight files; gating/repo upload date is not release date
Retrieved Oct 9, 2026 · factual metadata; model-specific licenses
Open source ↗ - License
- otherHugging Face HubPublished source fact
Retrieved Oct 9, 2026 · factual metadata; model-specific licenses
Open source ↗ - Input modalities
- textmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗
Benchmark results
2 results| Benchmark | Raw result | Normalized (0–100) | Pillar |
|---|---|---|---|
| swe bench verified | 71.3%Official model cards via models.devLab-reported; metric resolved; transcribed by MIT models.dev catalog; not independently evaluated [variant]
Retrieved Oct 9, 2026 · factual citation; MIT transcription Open source ↗ | 71.3 | coding |
| swe bench verified | 63.4%SWE-bench VerifiedSWE-bench published model plus agent result; harness retained, not a base model evaluation [variant] mini-SWE-agent; 1.17.2Published Dec 10, 2025
Retrieved Oct 9, 2026 · factual citation Open source ↗ | 63.4 | coding |
Normalization uses fixed absolute 0–100 scales for each unit, independently of other models that have a result on each benchmark. Coverage and evidence breadth still affect the composite; compare the evaluation conditions before reading a small score gap as decisive. “lab-reported” marks the provider's own published figure.
Sources: in and pending
in Reported (2)
- Official model cards via models.dev · arrived Oct 8, 2026
- SWE-bench Verified · arrived Oct 8, 2026
pending Awaiting (6)
- ARC Prize · carries 11% of expected weight
- Epoch AI Benchmarking · carries 23% of expected weight
- Humanity’s Last Exam · carries 11% of expected weight
- LiveBench · carries 11% of expected weight
- LMArena / Arena · carries 23% of expected weight
- Terminal-Bench · carries 6% of expected weight
The confidence % rises as pending sources publish. Some sources never cover some models — that is why 100% confidence arrives at 80% of expected weight, not at full coverage.