Loading benchmark catalogue
Compare performance benchmarks, coding intelligence, reasoning evaluations, context limits, and API economics between DeepSeek V4 Pro 0813 (DeepSeek) and Ling 3.1 Flash (InclusionAI).
| Metric / Capability | DeepSeek V4 Pro 0813 | Ling 3.1 Flash |
|---|---|---|
| Developer / Organization | DeepSeek | InclusionAI |
| Release Date | 2026-08-12 | 2026-10-02 |
| Deployment / Weights | Open Weights | Proprietary Cloud API |
| Context Window | 1.0M tokens (1,048,576 tokens) | 262K tokens (262,144 tokens) |
| Input Pricing (/1M Tokens) | Free (Open Weights) | $0.075 |
| Output Pricing (/1M Tokens) | Free (Open Weights) | $0.225 |
| Benchmark / Domain | DeepSeek V4 Pro 0813 | Ling 3.1 Flash | Advantage |
|---|---|---|---|
Artificial Analysis Intelligence Index Composite index aggregating 10+ evaluations of general reasoning and knowledge. | 36.0% | 41.1% | Ling 3.1 Flash (+14%) |
Artificial Analysis Coding Index Composite index of coding and software-engineering evaluations. | 68.8% | — | — |
Artificial Analysis Agentic Index Composite index of tool-use and long-horizon agentic tasks. | 41.3% | — | — |
SWE-bench Pro Harder, contamination-resistant successor to SWE-bench spanning multiple languages. | 55.4% | 68.7% | Ling 3.1 Flash (+24%) |
Terminal-Bench 4.0 Sandboxed command-line tasks requiring shell and file-system operation. | — | 46.0% | — |
ARC-AGI-2 Novel visual grid puzzles requiring abstract rule induction. | 61.3% | — | — |
GPQA Diamond Graduate-level physics, chemistry, and biology questions written by domain experts. | 90.1% | — | — |
Humanity's Last Exam Expert-written questions across hundreds of disciplines designed to resist saturation. | 42.7% | 37.4% | DeepSeek V4 Pro 0813 (+14%) |
FrontierMath v2 (Tiers 1-3) Unpublished, research-level mathematics problems across Tiers 1-3. | 45.3% | — | — |
Ling 3.1 Flash outperforms DeepSeek V4 Pro 0813 across 2 evaluated benchmarks compared to 1 for DeepSeek V4 Pro 0813, excelling in Artificial Analysis Intelligence Index, SWE-bench Pro.
DeepSeek V4 Pro 0813 offers a larger context capacity of 1.0M tokens (1,048,576 tokens), compared to 262K tokens for Ling 3.1 Flash.
DeepSeek V4 Pro 0813 is more cost-effective. Input tokens cost $Free (Open Weights)/M tokens vs $0.075/M tokens for Ling 3.1 Flash.
DeepSeek V4 Pro 0813 is an open-weights model. Ling 3.1 Flash is a proprietary cloud API.
Analyze 130+ models side-by-side using interactive radar charts, cost-efficiency quadrant plots, and live dataset filters on the full NeuBench index.
Open Interactive Comparator