Model comparison
Grok 4.6 vs Kimi K3
SpaceXAI against Moonshot AI, compared on context, price, and verified benchmark results.
Catalog record checked August 12, 2026; individual provider fields may change.
SpaceXAI
Grok 4.6
Frontier
Moonshot AI
Kimi K3
Frontier · Open weights
| Specification | Grok 4.6 | Kimi K3 |
|---|---|---|
| Provider | SpaceXAI | Moonshot AI |
| Tier | Frontier | Frontier |
| Context window | 500K | Winner: 1.05M |
| Max output | Not verified | 1M |
| Input / 1M tokens | Winner: $2 | $3 |
| Output / 1M tokens | Winner: $6 | $15 |
| Weights | Closed | Open |
| Parameters | Not disclosed | 2.8T total / 104B active (MoE) |
| Reasoning levels | low, medium, high, xhigh | low, high, max |
| Modalities | text, image | text, image, video |
| Released | August 12, 2026 | July 16, 2026 |
| Artificial Analysis Intelligence Index [high] (2026-08-12) | Winner: 61 | 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: Grok 4.6: Base tier is $2/$6 per million tokens; prompts at or above 200K tokens are priced at $4/$12. Cached input is $0.50 / $1.00. Web/X search tool calls bill separately.
Grok 4.6
SpaceXAI's current code and chat default — same $2/$6 list as Grok 4.5, with a 500K context window and image input.
Best for
- Coding agents
- Chat and knowledge work
- Cost-sensitive frontier work
Watch out
Token rates double above a 200K-token prompt. DeepSWE 1.1 best-effort for this family is the xhigh row at 66.7% (the catalog table rounds Pass@1 to 67%), not the higher-scoring medium config.
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
Both models 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
Grok 4.6 vs Kimi K3
Answered from the verified figures on this page rather than general guidance.
Is Grok 4.6 or Kimi K3 cheaper for input?
Grok 4.6 is cheaper at $2 per million input tokens, against $3 for Kimi K3 — roughly 1.5× the price. Output tokens usually dominate a real bill, so weigh the output rate more heavily than the input rate. These are base rates; Grok 4.6 has tiered pricing: Base tier is $2/$6 per million tokens; prompts at or above 200K tokens are priced at $4/$12. Cached input is $0.50 / $1.00. Web/X search tool calls bill separately.
Is Grok 4.6 or Kimi K3 cheaper for output?
Grok 4.6 is cheaper at $6 per million output tokens, against $15 for Kimi K3 — roughly 2.5× the price. Output tokens usually dominate a real bill, so weigh the output rate more heavily than the input rate. These are base rates; Grok 4.6 has tiered pricing: Base tier is $2/$6 per million tokens; prompts at or above 200K tokens are priced at $4/$12. Cached input is $0.50 / $1.00. Web/X search tool calls bill separately.
Which has the larger context window, Grok 4.6 or Kimi K3?
Kimi K3 accepts 1.05M tokens against 500K for Grok 4.6. This only matters if you routinely send very long documents or large codebases.
Do Grok 4.6 and Kimi K3 support the same reasoning levels?
Grok 4.6 exposes low, medium, high, xhigh, while Kimi K3 exposes low, high, max.
Should I use Grok 4.6 or Kimi K3?
Both sit in the frontier tier, so the choice usually comes down to price and context rather than capability. Grok 4.6 suits coding agents; Kimi K3 suits agentic coding without vendor lock-in.
Can I self-host Grok 4.6 or Kimi K3?
Kimi K3 publishes open weights, but self-hosting and third-party provider access depend on the licence, hardware, serving support, and availability. Grok 4.6 is a closed model whose supported access paths are controlled by its provider.
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.