Model comparison
Kimi K2.6 vs Kimi K3
Two Moonshot AI tiers compared on the figures that decide which one a workload actually needs.
Catalog record checked August 12, 2026; individual provider fields may change.
Moonshot AI
Kimi K2.6
Budget · Open weights
Moonshot AI
Kimi K3
Frontier · Open weights
| Specification | Kimi K2.6 | Kimi K3 |
|---|---|---|
| Provider | Moonshot AI | Moonshot AI |
| Tier | Budget | Frontier |
| Context window | 262K | Winner: 1.05M |
| Max output | Not verified | 1M |
| Input / 1M tokens | Winner: $0.95 | $3 |
| Output / 1M tokens | Winner: $4 | $15 |
| Weights | Open | Open |
| Parameters | 1T total / 32B active (MoE) | 2.8T total / 104B active (MoE) |
| Reasoning levels | Not verified | low, high, max |
| Modalities | text, image, video | text, image, video |
| Released | April 21, 2026 | July 16, 2026 |
| Artificial Analysis Intelligence Index [max] (2026-08-08) | 45 | Winner: 60 |
Prices are USD per million tokens at standard rates, excluding batch and caching discounts. Bold indicates the better figure where one is objectively better. Values we could not confirm from the provider are shown as “Not verified” rather than estimated.
Pricing tiers: Kimi K2.6: Cache-hit input is $0.16 / MTok; cache-miss input is $0.95 / MTok; output $4.00 / MTok (Moonshot list).
Kimi K2.6
Open-weight mixture-of-experts model priced well below the closed frontier tiers.
Best for
- Cost-sensitive volume
- Self-hosting
- Avoiding vendor lock-in
Watch out
262K context is among the smaller windows here — a real constraint on long-document work.
Kimi K3
Open-weight frontier model scoring within a few points of the closed leaders on agentic coding.
Best for
- Agentic coding without vendor lock-in
- Self-hosting at frontier quality
- Long-context work
Watch out
2.8T parameters means self-hosting is a datacentre exercise, not a workstation one — open weights here mean provider choice, not local inference.
Benchmark
DeepSWE 1.1 in context
Only Kimi K3 has a published DeepSWE 1.1 result. It is shown against the wider field, with cost per completed task alongside the score.
Local leader
Claude Opus 5 [max]
74%
Rows shown
24
Highest published reasoning effort per model (not best Pass@1)
Snapshot date
2026-08-13
Mirrored from deepswe.datacurve.ai
Better is toward the top-right (higher pass rate, lower cost). X-axis is reversed to match DeepSWE’s public chart. v1.1 uses average cost / tokens / steps; v1 uses published medians.
| # | Model | Pass@1 | Cost / task | Tokens / task | Steps / task |
|---|---|---|---|---|---|
| 1 | Claude Opus 5 [max] | 74% | $11.84 | 118k | 99 |
| 2 | GPT-5.6 Sol [max] | 73% | $8.39 | 60k | 61 |
| 3 | Claude Fable 5 [max] | 70% | $21.63 | 119k | 88 |
| 4 | GPT-5.6 Terra [max] | 70% | $4.95 | 72k | 76 |
| 5 | Kimi K3 [max] | 69% | $4.65 | 82k | 98 |
| 6 | GPT-5.6 Luna [max] | 67% | $3.03 | 73k | 102 |
| 7 | GPT-5.5 [xhigh] | 67% | $7.23 | 46k | 82 |
| 8 | Grok 4.6 [xhigh] | 67% | $5.50 | 71k | 87 |
| 9 | Gemini 3.7 Flash [high] | 65% | $2.18 | 107k | 125 |
| 10 | DeepSeek V4-Pro [max] | 63% | $0.24 | 106k | 155 |
| 11 | Claude Opus 4.8 [max] | 59% | $13.22 | 135k | 120 |
| 12 | Qwen3.8-Max [xhigh] | 58% | $3.73 | 95k | 111 |
| 13 | Muse Spark 1.2 [xhigh] | 55% | $3.70 | 99k | 101 |
| 14 | Claude Sonnet 5 [max] | 54% | $26.40 | 214k | 268 |
| 15 | Grok 4.5 [high] | 54% | $2.42 | 36k | 61 |
| 16 | DeepSeek V4-Flash [max] | 53% | $0.10 | 108k | 153 |
| 17 | Muse Spark 1.1 [xhigh] | 53% | $2.36 | 74k | 96 |
| 18 | GPT-5.4 [xhigh] | 52% | $5.65 | 71k | 70 |
| 19 | Gemini 3.6 Flash [high] | 47% | $4.42 | 96k | 117 |
| 20 | GLM 5.2 [max] | 44% | $3.92 | 78k | 129 |
| 21 | Gemini 3.5 Flash [high] | 36% | $3.45 | 76k | 105 |
| 22 | Kimi K2.7 Code | 31% | $2.82 | 59k | 149 |
| 23 | Claude Sonnet 4.6 [high] | 30% | $5.52 | 76k | 134 |
| 24 | Gemini 3.1 Pro [high] | 12% | $2.14 | 28k | 76 |
DeepSWE “Best” picks the highest published reasoning effort per model (not the highest pass rate). Small gaps may not be statistically meaningful — confirm on deepswe.datacurve.ai.
Common questions
Kimi K2.6 vs Kimi K3
Answered from the verified figures on this page rather than general guidance.
Is Kimi K2.6 or Kimi K3 cheaper for input?
Kimi K2.6 is cheaper at $0.95 per million input tokens, against $3 for Kimi K3 — roughly 3.2× the price. Output tokens usually dominate a real bill, so weigh the output rate more heavily than the input rate. These are base rates; Kimi K2.6 has tiered pricing: Cache-hit input is $0.16 / MTok; cache-miss input is $0.95 / MTok; output $4.00 / MTok (Moonshot list).
Is Kimi K2.6 or Kimi K3 cheaper for output?
Kimi K2.6 is cheaper at $4 per million output tokens, against $15 for Kimi K3 — roughly 3.8× the price. Output tokens usually dominate a real bill, so weigh the output rate more heavily than the input rate. These are base rates; Kimi K2.6 has tiered pricing: Cache-hit input is $0.16 / MTok; cache-miss input is $0.95 / MTok; output $4.00 / MTok (Moonshot list).
Which has the larger context window, Kimi K2.6 or Kimi K3?
Kimi K3 accepts 1.05M tokens against 262K for Kimi K2.6. This only matters if you routinely send very long documents or large codebases.
Should I use Kimi K2.6 or Kimi K3?
Kimi K2.6 is the budget tier and Kimi K3 the frontier tier. The useful question is whether your hardest task actually fails on the cheaper one — most production volume such as classification, extraction and summarisation does not.
Next step
Choosing between them
Tier and workload decide this more reliably than a leaderboard position does.
If both sit in the same tier, the decision usually comes down to context window and output price rather than headline capability — output tokens dominate real bills.
If one is a step up within the same provider, the useful question is whether your hardest task actually fails on the cheaper tier. Most production volume — classification, extraction, summarization — does not.
Choosing a harness rather than a model — Cursor, Copilot, Claude Code, Muse Code, or Lovable? Compare agentic harnesses · Latest releases.