Model comparison
Kimi K3 vs Step 3.7 Flash
Head-to-head evidence from 19 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Kimi K3 #4 (Supported); Step 3.7 Flash #110 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Kimi K3 and Step 3.7 Flash share 19 comparable benchmark results. 1 of 8 categories are comparable. 38 results are unique to Kimi K3; 10 to Step 3.7 Flash.
Updated July 23, 2026- Shared results
- 19
- Kimi K3 only
- 38
- Step 3.7 Flash only
- 10
- Comparable categories
- 1 / 8
Pick Kimi K3 if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 19 shared benchmark results across 5 evidence categories; 1 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Kimi K3 is clearly ahead on the BenchAlign aggregate, 80.96 to 50.87. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Kimi K3's sharpest advantage is in agentic, where it averages 89.5 against 66.4. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 88.3% to 59.5%.
Kimi K3 is also the more expensive model on tokens at $3.00 input / $15.00 output per 1M tokens, versus $0.20 input / $1.15 output per 1M tokens for Step 3.7 Flash. That is roughly 13.0x on output cost alone. Kimi K3 gives you the larger context window at 1.05M, compared with 256K for Step 3.7 Flash.
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 | Kimi K3 | Δ | Step 3.7 Flash |
|---|---|---|---|
| Agentic | Kimi K389.5 | Margin← 23.1 | Step 3.7 Flash66.4 |
| Coding | Kimi K3Not measured | MarginNo overlap | Step 3.7 Flash56.3 |
| Knowledge | Kimi K361.0 | MarginNo overlap | Step 3.7 FlashNot measured |
| Multimodal | Kimi K378.5 | MarginNo overlap | Step 3.7 FlashNot measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 88.3%B 59.5%Winner: Kimi K3Δ 28.8Terminal-Bench 2.0: Kimi K3 scored 88.3%; Step 3.7 Flash scored 59.5%. Kimi K3 wins this benchmark. - Source ↗
BrowseComp
AgenticA 91.2%B 75.8%Winner: Kimi K3Δ 15.4BrowseComp: Kimi K3 scored 91.2%; Step 3.7 Flash scored 75.8%. Kimi K3 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Kimi K3 | Step 3.7 Flash | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Kimi K3$3 input / $15 output | Step 3.7 Flash$0.2 input / $1.15 output | Step 3.7 Flash has the lower combined listed price. |
| Generation speedtokens per second | Kimi K3Not available | Step 3.7 FlashNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Kimi K3Not available | Step 3.7 FlashNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Kimi K31.05M | Step 3.7 Flash256K | Kimi K3 lists the larger context window. |
Benchmark Deep Dive
AgenticKimi K3 wins26 benchmarks
| Benchmark | Kimi K3 | Step 3.7 Flash | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 88.3% | 59.5% | Kimi K3 leads |
| BrowseCompSource | 91.2% | 75.8% | Kimi K3 leads |
| DeepSearchQASource | 95.0% | 92.8% | Kimi K3 leads |
| 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 Agentic IndexSource | 50.1% | 21.5% | Kimi K3 leads |
| GDPval-AASource | 59.0% | 25.9% | Kimi K3 leads |
| GDPval-AASource | 1679 | 1017 | Kimi K3 leads |
| 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% | 14.8% | Kimi K3 leads |
| AA ITBenchSource | 47.7% | — | Not comparable |
| ToolathlonSource | — | 49.5% | Not comparable |
| Claw-EvalSource | — | 67.1% | Not comparable |
| HLE w/ toolsSource | — | 47.2% | Not comparable |
| Gert LabsSource | — | 51.57% | Not comparable |
| τ²-bench resultsSource | — | 98.5% | Not comparable |
Coding11 benchmarks
| Benchmark | Kimi K3 | Step 3.7 Flash | Result |
|---|---|---|---|
| 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 |
| AA Coding IndexSource | 76.2% | 39.6% | Kimi K3 leads |
| AA-SciCodeSource | 58.7% | 40.0% | Kimi K3 leads |
| SWE-bench ProSource | — | 56.3% | Not comparable |
| Terminal-Bench 2.0Source | — | 59.5% | Not comparable |
Reasoning2 benchmarks
Knowledge10 benchmarks
| Benchmark | Kimi K3 | Step 3.7 Flash | Result |
|---|---|---|---|
| GPQASource | 93.5% | — | Not comparable |
| GPQA-DSource | 93.5% | — | Not comparable |
| HLESource | 56% | — | Not comparable |
| HLE w/o toolsSource | 43.5% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 57.1% | 30.3% | Kimi K3 leads |
| AA-GPQA DiamondSource | 93.5% | 80.9% | Kimi K3 leads |
| AA-HLESource | 44.3% | 19.9% | Kimi K3 leads |
| AA-Omniscience IndexSource | 18.4% | -37.5% | Kimi K3 leads |
| AA-Omniscience AccuracySource | 46.0% | 25.4% | Kimi K3 leads |
| AA-Omniscience Hallucination RateSource | 50.9% | 84.4% | Kimi K3 leads |
Multimodal17 benchmarks
| Benchmark | Kimi K3 | Step 3.7 Flash | 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% | 75.3% | Kimi K3 leads |
| Design Arena WebsiteSource | 1386 | 1211 | Kimi K3 leads |
| SimpleVQASource | — | 79.2% | Not comparable |
| V*Source | — | 95.3% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Kimi K3 | Step 3.7 Flash | Result |
|---|---|---|---|
| AA-IFBenchSource | — | 67.3% | Not comparable |
Frequently Asked Questions (2)
Which is better, Kimi K3 or Step 3.7 Flash?
Kimi K3 is ahead on BenchLM's BenchAlign leaderboard, 80.96 to 50.87. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 88.3% and 59.5%.
Which is better for agentic tasks, Kimi K3 or Step 3.7 Flash?
Kimi K3 has the edge for agentic tasks in this comparison, averaging 89.5 versus 66.4. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Related Comparisons
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.