Kimi K3 vs Hy4 preview
At a Glance
| Compare |  Kimi K3Moonshot AI |  Hy4 previewTencent |
|---|---|---|
| Intelligence, Cost, and Efficiency | ||
| IntelligenceHigher is better · MM Intelligence v4.4 | #13 of 5773.3 score · 3/3 sources · complete | UnrankedNot in the 57-model eligible cohort |
| CostLower is better · Published-token output estimate | #19 of 27$0.163 per LiveBench case | UnrankedNot in the 27-model eligible cohort |
| EfficiencyHigher is better · MM Efficiency v1.9 | #12 of 2750.1 score · 3/3 sources · complete | UnrankedNot in the 27-model eligible cohort |
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.239Openrouter ↗ · Oct 19, 2026 | $0.834Deepinfra ↗ · Oct 19, 2026 |
| Output priceFrom · USD / 1M tokens | $11.936Openrouter ↗ · Oct 19, 2026 | $2.501Deepinfra ↗ · Oct 19, 2026 |
| Context windowMaximum documented tokens | 1,049K | 1,000K |
| Model facts checked | Aug 28, 2026View model evidence → | Sep 2, 2026View model evidence → |
Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →
Available Benchmarks
| Benchmark | Kimi-K3 | Hy4 preview |
|---|---|---|
| LMArena Agent Arenaagent-2026-10-16-a64ca3b12b6e · outcome_score · statistical tie | 1.9399% of row best · score · Kimi K3 (Max); 95% CI [1.20183745, 2.66737332]; sessions 135415; observations 12343129; rank 23 | 2.68100% of row best · score · Hy4 preview; 95% CI [2.00929922, 3.35139547]; sessions 84828; observations 8314177; rank 20 |
| Overall ResultCounted from the protocol-matched rows above · 1 tie | 0 benchmark winsNo overall winner | 0 benchmark winsNo overall winner |
Third-party benchmark Only like-for-like primary-publisher results are shown; raw scores, relative scores, configuration, and token spend remain visible.
Side-by-Side Facts
| Field | Kimi-K3 | Hy4 preview |
|---|---|---|
| Developer | Moonshot AI | Tencent |
| Family | Kimi K3 | Hy4 |
| Model | Kimi-K3 | Hy4 preview |
| Version | Kimi-K3 | Hy4 preview |
| Lifecycle | active | preview |
| Released | 2026-07-16 | 2026-08-28 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Context window | 1,049K | 1,000K |
| Total parameters | 2.78T | 770B |
| Active parameters | 104B | 49B |
| License | other | apache-2.0 |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Deepinfra (Standard), Fireworks Ai (Standard), Hugging Face (Standard), Openrouter (Standard), Together Ai (Standard) | Deepinfra (Standard), Openrouter (Standard) |
| Capabilities | chat, generation, reasoning | chat, generation, reasoning, tools |
Kimi K3 Capabilities
Hy4 preview Capabilities
Related Comparisons
All text comparisons →| A | Pair | B | Context |
|---|---|---|---|
Kimi-K2.6Moonshot AI | vs | Kimi-K3Moonshot AI | family variantsembedding, image, text |
Claude Fable 5.1Anthropic | vs | Kimi-K3Moonshot AI | cross-developer peersimage, text |
Claude Mythos 5.1Anthropic | vs | Kimi-K3Moonshot AI | cross-developer peersimage, text |
Claude Opus 5.5Anthropic | vs | Kimi-K3Moonshot AI | cross-developer peersimage, text |
 DeepSeek-V4-ProDeepSeek | vs | Kimi-K3Moonshot AI | cross-developer peerstext |
DeepSeek-V4-Pro-BaseDeepSeek | vs | Kimi-K3Moonshot AI | cross-developer peerstext |
Kimi-K3Moonshot AI | vs |  GPT-6.1 SolOpenAI | cross-developer peersimage, text |
Kimi-K3Moonshot AI | vs |  MiMo-V2.6-ProXiaomi | cross-developer peersimage, text |
DeepSeek-V4-Pro-0813DeepSeek | vs | Kimi-K3Moonshot AI | cross-developer peerstext |
 Mistral-Small-4-119B-2603Mistral AI | vs | Kimi-K3Moonshot AI | cross-developer peersimage, text |
Kimi-K2-ThinkingMoonshot AI | vs | Kimi-K3Moonshot AI | family variantstext |
Kimi-K2.5Moonshot AI | vs | Kimi-K3Moonshot AI | family variantsembedding, image, text |
Primary Evidence
Sources and Freshness
Questions
Kimi K3 vs Hy4 preview FAQs
Is Kimi K3 or Hy4 preview better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both Kimi K3 and Hy4 preview, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, Kimi K3 or Hy4 preview?+
Kimi K3 is $0.239 and Hy4 preview is $0.834 per million tokens, so Kimi K3 is cheaper on this metric. Kimi K3 is $11.936 and Hy4 preview is $2.501 per million tokens, so Hy4 preview is cheaper on this metric.
Which has a larger context window, Kimi K3 or Hy4 preview?+
Kimi K3 has the larger sourced context window. Kimi K3 supports 1,048,576 tokens and Hy4 preview supports 1,000,000 tokens.
Which performs better in benchmarks, Kimi K3 or Hy4 preview?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can Kimi K3 or Hy4 preview be self-hosted?+
Both models have the same sourced self-hosting status: supported. Kimi K3 is open weight; Hy4 preview is open weight.
Can Kimi K3 and Hy4 preview understand images?+
Kimi K3 is documented with image input; Hy4 preview is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, Kimi K3 or Hy4 preview?+
Neither model has a sourced maximum output in this comparison, so a larger limit cannot be determined.
Do Kimi K3 and Hy4 preview support reasoning and tool use?+
Kimi K3: reasoning and image input. Hy4 preview: reasoning and tool calling. Feature support does not establish relative quality.
Which is available from more inference providers, Kimi K3 or Hy4 preview?+
Kimi K3 has 5 sourced provider routes; Hy4 preview has 2, so Kimi K3 has broader tracked availability.
Which offers better value, Kimi K3 or Hy4 preview?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.