# DeepSeek V4 Pro 0813 vs GPT-6 Astra

> Side-by-side benchmark comparison for DeepSeek V4 Pro 0813 and GPT-6 Astra across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/deepseek-v4-pro-0813-vs-gpt-6-astra
- Last updated: October 17, 2026

- Shared sourced benchmarks: 10
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

GPT-6 Astra leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- GPT-6 Astra has the higher overall point estimate 83.88 to 63.19. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in reasoning, where the averages are 60.7 for DeepSeek V4 Pro 0813 and 93.3 for GPT-6 Astra.
- The biggest single benchmark swing is ARC-AGI-2 in Reasoning, with scores of 61.3% and 95%.
- DeepSeek V4 Pro 0813 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-6 Astra also has the larger context window at 1.05M.

## Model Snapshot

| Property | DeepSeek V4 Pro 0813 | GPT-6 Astra |
|----------|----------|----------|
| Creator | DeepSeek | OpenAI |
| Type | Open Weight | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | 1M | 1.05M |
| Overall Score | 63.19 | 83.88 |
| Benchmarks Covered | 41 | 40 |
| Pricing (input/output) | $1.32 / $3.96 | $10.00 / $50.00 |

## Category Breakdown

### Agentic

- Winner: GPT-6 Astra
- DeepSeek V4 Pro 0813 public-lane score: 54.5 (Supported · #43/128)
- GPT-6 Astra public-lane score: 74.8 (Supported · #7/128)

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 67.9% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 87.9% | Coming soon | Coming soon |
| BrowseComp | 83.4% | 91.5% | GPT-6 Astra |
| HLE w/ tools | 60.0% | 57.2% | DeepSeek V4 Pro 0813 |
| MCP Atlas | 73.6% | Coming soon | Coming soon |
| Toolathlon | 51.8% | Coming soon | Coming soon |
| CyberGym | 83.3% | Coming soon | Coming soon |
| Toolathlon-Verified | 74.1% | Coming soon | Coming soon |
| Agents' Last Exam | 25.7% | 59.3% | GPT-6 Astra |
| AutomationBench | 31.8% | 41.4% | GPT-6 Astra |
| Terminal-Bench 2.1 (Vals) | 54.7% | 87.3% | GPT-6 Astra |
| OSWorld 2.0 | Coming soon | 72.6% | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 57.90% | Coming soon |
| Terminal-Bench-Science 0.1 | Coming soon | 64.6% | Coming soon |
| ExploitGym | Coming soon | 42.4% | Coming soon |
| ApprenticeBench | Coming soon | 68% | Coming soon |
| CWE-bench v1 | Coming soon | 68.0% | Coming soon |
| Hermes Index | Coming soon | 56.3% | Coming soon |
| Hermes Bench | Coming soon | 73.7% | Coming soon |
| MazeBench | Coming soon | 23% | Coming soon |

### Coding

- Winner: GPT-6 Astra
- DeepSeek V4 Pro 0813 public-lane score: 47.2 (Supported · #49/150)
- GPT-6 Astra public-lane score: 75.5 (Supported · #4/150)

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench Pass@1-COT | 93.5% | Coming soon | Coming soon |
| Codeforces | 3206.0 | Coming soon | Coming soon |
| SWE-bench Verified | 80.6% | Coming soon | Coming soon |
| SWE-bench Pro | 55.4% | Coming soon | Coming soon |
| SWE Multilingual | 76.2% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 67.9% | Coming soon | Coming soon |
| Vibe Code Bench | 49.93% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 87.9% | Coming soon | Coming soon |
| NL2Repo | 61.5% | Coming soon | Coming soon |
| DeepSWE | 62.7% | 74.1% | GPT-6 Astra |
| DSBench-FullStack | 71.1% | Coming soon | Coming soon |
| DSBench-Hard | 67.2% | Coming soon | Coming soon |
| OpenHarmony Bench | 59.0% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 87.5% | Coming soon | Coming soon |
| SWE-bench (Vals) | 96.4% | Coming soon | Coming soon |
| FrontierCode 1.1 Main | Coming soon | 53.3% | Coming soon |
| FrontierCode 1.1 Extended | Coming soon | 64.5% | Coming soon |
| FrontierSWE v2 | Coming soon | 65.5% | Coming soon |
| PostTrainBench v1.1 | Coming soon | 44.3% | Coming soon |
| Bug Hunt Bench | Coming soon | 45.0 fixes | Coming soon |

### Reasoning

- Winner: Not comparable
- DeepSeek V4 Pro 0813 public-lane score: 60.7 (Unranked · 4 rankable rows)
- GPT-6 Astra public-lane score: 93.3 (#1/28)

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| MRCR 1M | 83.5% | Coming soon | Coming soon |
| CorpusQA 1M | 62.0% | Coming soon | Coming soon |
| ARC-AGI-1 | 90.00% | 98.50% | GPT-6 Astra |
| ARC-AGI-2 | 61.3% | 95% | GPT-6 Astra |
| STEMBench | Coming soon | 2.81 | Coming soon |
| ARC-AGI-3 | Coming soon | 62.7% | Coming soon |
| GeneBench-Pro | Coming soon | 37.8% | Coming soon |
| MRCR v2 256K-512K | Coming soon | 100.0% | Coming soon |
| MRCR v2 512K-1M | Coming soon | 96.3% | Coming soon |
| GraphWalks BFS 256K–1M | Coming soon | 71.8% | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- DeepSeek V4 Pro 0813 public-lane score: Coming soon
- GPT-6 Astra public-lane score: 87.3 (Unranked · 3 rankable rows)

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| ScreenSpot Pro | Coming soon | 92.7% | Coming soon |
| BenchCAD Vision2Code (tools) | Coming soon | 0.959 | Coming soon |

### Knowledge

- Winner: Directional only
- DeepSeek V4 Pro 0813 public-lane score: 63.5 (Estimated · #41/184)
- GPT-6 Astra public-lane score: 86.2 (Supported · #3/184)

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| MMLU-Pro | 87.5% | Coming soon | Coming soon |
| Chinese-SimpleQA | 84.4% | Coming soon | Coming soon |
| GPQA | 90.1% | 96% | GPT-6 Astra |
| GPQA-D | 90.1% | 96.0% | GPT-6 Astra |
| HLE | 42.7% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 92.4% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 87.0% | Coming soon | Coming soon |
| HealthBench (raw) | Coming soon | 56.9% | Coming soon |
| HealthBench (length-adjusted) | Coming soon | 58.3% | Coming soon |
| HealthBench Professional | Coming soon | 64.7% | Coming soon |
| HealthBench Professional (raw) | Coming soon | 68.2% | Coming soon |
| HealthBench Hard | Coming soon | 36.6% | Coming soon |

### Instruction Following

- Winner: Not comparable
- DeepSeek V4 Pro 0813 public-lane score: Coming soon
- GPT-6 Astra public-lane score: Coming soon

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| Gray Swan IPI (15 attempts) | Coming soon | 8.5% | Coming soon |

### Mathematics

- Winner: Not comparable
- DeepSeek V4 Pro 0813 public-lane score: 87.4 (Unranked · 4 rankable rows)
- GPT-6 Astra public-lane score: 97.3 (Unranked · 2 rankable rows)

| Benchmark | DeepSeek V4 Pro 0813 | GPT-6 Astra | Winner |
|-----------|-----------|-----------|--------|
| HMMT Feb 2026 | 95.2% | Coming soon | Coming soon |
| IMOAnswerBench | 89.8% | Coming soon | Coming soon |
| Apex | 38.3% | Coming soon | Coming soon |
| Apex Shortlist | 90.2% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | Coming soon | 97.600% | Coming soon |
| FrontierMath v2 (Tiers 1-3) | Coming soon | 93.684% | Coming soon |

## FAQ

### Which is better overall, DeepSeek V4 Pro 0813 or GPT-6 Astra?

GPT-6 Astra has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### Where is the biggest gap between DeepSeek V4 Pro 0813 and GPT-6 Astra?

The widest category gap is in reasoning, where the averages are 60.7 for DeepSeek V4 Pro 0813 and 93.3 for GPT-6 Astra.

### How many benchmarks does DeepSeek V4 Pro 0813 cover on BenchLM?

DeepSeek V4 Pro 0813 currently has 41 sourced benchmark scores on BenchLM.

### How many benchmarks does GPT-6 Astra cover on BenchLM?

GPT-6 Astra currently has 40 sourced benchmark scores on BenchLM.

## Related Comparisons

- [DeepSeek V4 Pro 0813 vs GPT-6.1 Sol](/compare/deepseek-v4-pro-0813-vs-gpt-6-1-sol)
- [GPT-6 Astra vs GPT-6.1 Sol](/compare/gpt-6-1-sol-vs-gpt-6-astra)
- [DeepSeek V4 Pro 0813 vs GPT-6 Sol](/compare/deepseek-v4-pro-0813-vs-gpt-6-sol)
- [GPT-6 Astra vs GPT-6 Sol](/compare/gpt-6-astra-vs-gpt-6-sol)

## Explore More

- [DeepSeek V4 Pro 0813 profile](/models/deepseek-v4-pro-0813)
- [GPT-6 Astra profile](/models/gpt-6-astra)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)