Model comparison
Hy3 Preview vs Kimi K3
Head-to-head evidence from 17 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Hy3 Preview unranked (Not scored); Kimi K3 #4 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Hy3 Preview and Kimi K3 share 17 comparable benchmark results. 2 of 8 categories are comparable. 5 results are unique to Hy3 Preview; 40 to Kimi K3.
Updated July 23, 2026- Shared results
- 17
- Hy3 Preview only
- 5
- Kimi K3 only
- 40
- Comparable categories
- 2 / 8
Treat this as a split decision. Hy3 Preview makes more sense if you want the cheaper token bill; Kimi K3 is the better fit if agentic is the priority or you need the larger 1.05M context window.
Confidence note. This is a partial-evidence comparison with 17 shared benchmark results across 4 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Hy3 Preview and Kimi K3 finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.
Kimi K3 is also the more expensive model on tokens at $3.00 input / $15.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Hy3 Preview. That is roughly Infinityx on output cost alone. Kimi K3 gives you the larger context window at 1.05M, compared with 256K for Hy3 Preview.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Hy3 Preview | Δ | Kimi K3 |
|---|---|---|---|
| Agentic | Hy3 Preview54.4 | Margin→ 35.1 | Kimi K389.5 |
| Knowledge | Hy3 Preview33.8 | Margin→ 27.2 | Kimi K361.0 |
| Coding | Hy3 Preview57.8 | MarginNo overlap | Kimi K3Not measured |
| Multimodal | Hy3 PreviewNot measured | MarginNo overlap | Kimi K378.5 |
| Inst. Following | Hy3 Preview63.1 | MarginNo overlap | Kimi K3Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 54.4%B 88.3%Winner: Kimi K3Δ 33.9Terminal-Bench 2.0: Hy3 Preview scored 54.4%; Kimi K3 scored 88.3%. Kimi K3 wins this benchmark. - Source ↗
HLE
KnowledgeA 25.5%B 56%Winner: Kimi K3Δ 30.5HLE: Hy3 Preview scored 25.5%; Kimi K3 scored 56%. Kimi K3 wins this benchmark. - Source ↗
GPQA
KnowledgeA 87.2%B 93.5%Winner: Kimi K3Δ 6.3GPQA: Hy3 Preview scored 87.2%; Kimi K3 scored 93.5%. Kimi K3 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Hy3 Preview | Kimi K3 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Hy3 Preview$0 input / $0 output | Kimi K3$3 input / $15 output | Hy3 Preview has the lower combined listed price. |
| Generation speedtokens per second | Hy3 PreviewNot available | Kimi K3Not available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Hy3 PreviewNot available | Kimi K3Not available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Hy3 Preview256K | Kimi K31.05M | Kimi K3 lists the larger context window. |
Benchmark Deep Dive
AgenticKimi K3 wins22 benchmarks
| Benchmark | Hy3 Preview | Kimi K3 | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 54.4% | 88.3% | Kimi K3 leads |
| Gert LabsSource | 36.91% | — | Not comparable |
| AA Agentic IndexSource | 30.7% | 50.1% | Kimi K3 leads |
| GDPval-AASource | 35.7% | 59.0% | Kimi K3 leads |
| GDPval-AASource | 1214 | 1679 | Kimi K3 leads |
| BrowseCompSource | — | 91.2% | Not comparable |
| DeepSearchQASource | — | 95.0% | Not comparable |
| Toolathlon-VerifiedSource | — | 73.2% | Not comparable |
| MCP AtlasSource | — | 84.2% | Not comparable |
| AutomationBenchSource | — | 30.8% | Not comparable |
| JobBenchSource | — | 52.9% | Not comparable |
| APEX-AgentsSource | — | 37.6% | Not comparable |
| SpreadsheetBench 2Source | — | 34.8% | Not comparable |
| DECK-BenchSource | — | 73.5% | Not comparable |
| AA BriefcaseSource | — | 1543 | Not comparable |
| AA AutomationBenchSource | — | 52.7% | Not comparable |
| AA EnterpriseOps-GymSource | — | 45.3% | Not comparable |
| AA Harvey LABSource | — | 94.6% | Not comparable |
| AA Tau3 BankingSource | — | 33.4% | Not comparable |
| aaTerminalBench21Source | — | 85% | Not comparable |
| APEX-Agents-AASource | — | 41.3% | Not comparable |
| AA ITBenchSource | — | 47.7% | Not comparable |
Coding12 benchmarks
| Benchmark | Hy3 Preview | Kimi K3 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 74.4% | — | Not comparable |
| Terminal-Bench 2.0Source | 54.4% | — | Not comparable |
| SciCodeSource | 41.2% | — | Not comparable |
| AA-SciCodeSource | 47.6% | 58.7% | Kimi K3 leads |
| AA Coding IndexSource | 58.8% | 76.2% | Kimi K3 leads |
| deepSweSource | — | 67.5% | Not comparable |
| FrontierSWESource | — | 81.2% | Not comparable |
| ProgramBenchSource | — | 77.8% | Not comparable |
| Kimi Code Bench v2Source | — | 72.9% | Not comparable |
| sweMarathonSource | — | 42% | Not comparable |
| PostTrain BenchSource | — | 36.6% | Not comparable |
| MLS-Bench LiteSource | — | 48.3% | Not comparable |
Reasoning2 benchmarks
KnowledgeKimi K3 wins10 benchmarks
| Benchmark | Hy3 Preview | Kimi K3 | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 41.2% | 57.1% | Kimi K3 leads |
| GPQASource | 87.2% | 93.5% | Kimi K3 leads |
| GPQA-DSource | 87.2% | 93.5% | Kimi K3 leads |
| HLESource | 25.5% | 56% | Kimi K3 leads |
| AA-Omniscience AccuracySource | 31.5% | 46.0% | Kimi K3 leads |
| AA-Omniscience Hallucination RateSource | 73.0% | 50.9% | Kimi K3 leads |
| AA-GPQA DiamondSource | 89.7% | 93.5% | Kimi K3 leads |
| AA-HLESource | 31.6% | 44.3% | Kimi K3 leads |
| AA-Omniscience IndexSource | -18.5% | 18.4% | Kimi K3 leads |
| HLE w/o toolsSource | — | 43.5% | Not comparable |
Multimodal15 benchmarks
| Benchmark | Hy3 Preview | Kimi K3 | Result |
|---|---|---|---|
| OfficeQA ProSource | — | 63.3% | Not comparable |
| MMMU-ProSource | — | 81.6% | Not comparable |
| MMMU-Pro w/ PythonSource | — | 83.4% | Not comparable |
| CharXiv w/o toolsSource | — | 84.8% | Not comparable |
| CharXivSource | — | 91.3% | Not comparable |
| MathVisionSource | — | 94.3% | Not comparable |
| MathVision w/ PythonSource | — | 97.8% | Not comparable |
| BabyVision w/ PythonSource | — | 85.7% | Not comparable |
| ZeroBenchSource | — | 23.0% | Not comparable |
| ZeroBench w/ PythonSource | — | 41.0% | Not comparable |
| WorldVQA ForceAnswerSource | — | 51.0% | Not comparable |
| OmniDocBenchSource | — | 91.1% | Not comparable |
| PerceptionBenchSource | — | 58.5% | Not comparable |
| AA-MMMU-ProSource | — | 80.5% | Not comparable |
| Design Arena WebsiteSource | — | 1386 | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Hy3 Preview | Kimi K3 | Result |
|---|---|---|---|
| IFBenchSource | 63.1% | — | Not comparable |
Frequently Asked Questions (3)
Which is better, Hy3 Preview or Kimi K3?
Hy3 Preview and Kimi K3 are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.
Which is better for knowledge tasks, Hy3 Preview or Kimi K3?
Kimi K3 has the edge for knowledge tasks in this comparison, averaging 61 versus 33.8. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Hy3 Preview or Kimi K3?
Kimi K3 has the edge for agentic tasks in this comparison, averaging 89.5 versus 54.4. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.