Model comparison
Ling 2.6 Flash vs Step 3.7 Flash
Head-to-head evidence from 16 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
BenchAlign evidence: Ling 2.6 Flash estimated; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Ling 2.6 Flash and Step 3.7 Flash share 16 comparable benchmark results. 1 of 8 categories are comparable. 3 results are unique to Ling 2.6 Flash; 14 to Step 3.7 Flash.
Updated July 16, 2026- Shared results
- 16
- Ling 2.6 Flash only
- 3
- Step 3.7 Flash only
- 14
- Comparable categories
- 1 / 8
Pick Step 3.7 Flash if you want the stronger benchmark profile. Ling 2.6 Flash only becomes the better choice if you need the larger 262K context window or you would rather avoid the extra latency and token burn of a reasoning model.
Confidence note. This is a partial-evidence comparison with 16 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
Step 3.7 Flash is clearly ahead on the provisional aggregate, 57 to 36. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Step 3.7 Flash's sharpest advantage is in coding, where it averages 56.3 against 27.
Step 3.7 Flash is the reasoning model in the pair, while Ling 2.6 Flash is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Ling 2.6 Flash gives you the larger context window at 262K, 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 | Ling 2.6 Flash | Δ | Step 3.7 Flash |
|---|---|---|---|
| Coding | Ling 2.6 Flash27.0 | Margin→ 29.3 | Step 3.7 Flash56.3 |
| Agentic | Ling 2.6 FlashNot measured | MarginNo overlap | Step 3.7 Flash66.4 |
| Knowledge | Ling 2.6 Flash59.0 | MarginNo overlap | Step 3.7 FlashNot measured |
| Inst. Following | Ling 2.6 Flash57.0 | MarginNo overlap | Step 3.7 FlashNot measured |
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Ling 2.6 Flash | Step 3.7 Flash | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Ling 2.6 FlashNot available | Step 3.7 Flash$0.2 input / $1.15 output | A complete price comparison is not available. |
| Generation speedtokens per second | Ling 2.6 Flash209.5 tok/s | Step 3.7 FlashNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Ling 2.6 Flash1.07 s | Step 3.7 FlashNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Ling 2.6 Flash262K | Step 3.7 Flash256K | Ling 2.6 Flash lists the larger context window. |
Benchmark Deep Dive
Agentic12 benchmarks
| Benchmark | Ling 2.6 Flash | Step 3.7 Flash | Result |
|---|---|---|---|
| τ²-bench resultsSource | 86% | 98.5% | Step 3.7 Flash leads |
| GDPval-AASource | 2.2% | 25.9% | Step 3.7 Flash leads |
| GDPval-AASource | 545 | 1017 | Step 3.7 Flash leads |
| AA Agentic IndexSource | 2.3% | 21.5% | Step 3.7 Flash leads |
| Terminal-Bench 2.0Source | — | 59.5% | Not comparable |
| BrowseCompSource | — | 75.8% | Not comparable |
| DeepSearchQASource | — | 92.8% | 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 |
| APEX-Agents-AASource | — | 14.8% | Not comparable |
CodingStep 3.7 Flash wins6 benchmarks
| Benchmark | Ling 2.6 Flash | Step 3.7 Flash | Result |
|---|---|---|---|
| SciCodeSource | 27% | — | Not comparable |
| AA Coding IndexSource | 25.3% | 39.6% | Step 3.7 Flash leads |
| Terminal-Bench HardSource | 21.2% | 35.6% | Step 3.7 Flash leads |
| AA-SciCodeSource | 27.1% | 40.0% | Step 3.7 Flash leads |
| SWE-bench ProSource | — | 56.3% | Not comparable |
| Terminal-Bench 2.0Source | — | 59.5% | Not comparable |
Reasoning2 benchmarks
Knowledge7 benchmarks
| Benchmark | Ling 2.6 Flash | Step 3.7 Flash | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 14.1% | 30.3% | Step 3.7 Flash leads |
| GPQASource | 59% | — | Not comparable |
| AA-GPQA DiamondSource | 59.3% | 80.9% | Step 3.7 Flash leads |
| AA-HLESource | 6.2% | 19.9% | Step 3.7 Flash leads |
| AA-Omniscience IndexSource | -65.7% | -37.5% | Step 3.7 Flash leads |
| AA-Omniscience AccuracySource | 15.4% | 25.4% | Step 3.7 Flash leads |
| AA-Omniscience Hallucination RateSource | 95.8% | 84.4% | Step 3.7 Flash leads |
Multimodal4 benchmarks
Frequently Asked Questions (2)
Which is better, Ling 2.6 Flash or Step 3.7 Flash?
Step 3.7 Flash is ahead on BenchLM's provisional leaderboard, 57 to 36.
Which is better for coding, Ling 2.6 Flash or Step 3.7 Flash?
Step 3.7 Flash has the edge for coding in this comparison, averaging 56.3 versus 27. Inside this category, Terminal-Bench Hard is the benchmark that creates the most daylight between them.
Related Comparisons
The AI models change fast. We track them for you.
A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.