# GPT-5.6 Terra vs Beam 1.0 501B-A23B NVFP4

> Side-by-side benchmark comparison for GPT-5.6 Terra and Beam 1.0 501B-A23B NVFP4 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-6-terra-vs-reflection-beam-nvfp4
- Last updated: October 16, 2026

- Shared sourced benchmarks: 6
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Pick GPT-5.6 Terra if you want the stronger benchmark profile. Beam 1.0 501B-A23B NVFP4 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GPT-5.6 Terra leads overall 70.19 to 70.
- The clearest category separation is in reasoning, where the averages are 69.1 for GPT-5.6 Terra and 85.5 for Beam 1.0 501B-A23B NVFP4.
- The biggest single benchmark swing is DeepSWE in Coding, with scores of 69.6% and 42.7%.
- Beam 1.0 501B-A23B NVFP4 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-5.6 Terra also has the larger context window at 1.05M.

## Model Snapshot

| Property | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 |
|----------|----------|----------|
| Creator | OpenAI | Reflection AI |
| Type | Proprietary | Open Weight |
| Reasoning | Reasoning | Reasoning |
| Context | 1.05M | 1M |
| Overall Score | 70.19 | null |
| Benchmarks Covered | 35 | 19 |
| Pricing (input/output) | $2.00 / $12.00 | Pricing unavailable / Pricing unavailable |

## Category Breakdown

### Agentic

- Winner: Not comparable
- GPT-5.6 Terra public-lane score: 60.4 (Supported · #26/128)
- Beam 1.0 501B-A23B NVFP4 public-lane score: Coming soon

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 3.0 | 20.8% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 87.4% | 78.4% | GPT-5.6 Terra |
| BrowseComp | 87.5% | 75.6% | GPT-5.6 Terra |
| OSWorld 2.0 | 50.2% | Coming soon | Coming soon |
| CyberGym | 81.8% | Coming soon | Coming soon |
| ExploitGym | 23.2% | Coming soon | Coming soon |
| Toolathlon | 53.1% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 77.5% | Coming soon | Coming soon |
| ApprenticeBench | 16% | Coming soon | Coming soon |
| MazeBench | 6% | Coming soon | Coming soon |
| τ³-bench results | Coming soon | 35.4% | Coming soon |
| HLE w/ tools | Coming soon | 51.1% | Coming soon |
| MCP Atlas | Coming soon | 80.5% | Coming soon |
| DeepSearchQA | Coming soon | 81.1% | Coming soon |

### Coding

- Winner: Not comparable
- GPT-5.6 Terra public-lane score: 61.9 (Supported · #15/150)
- Beam 1.0 501B-A23B NVFP4 public-lane score: Coming soon

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Pro | 63.4% | 65.6% | Beam 1.0 501B-A23B NVFP4 |
| Terminal-Bench 2.1 | 87.4% | 78.4% | GPT-5.6 Terra |
| DeepSWE | 69.6% | 42.7% | GPT-5.6 Terra |
| FrontierCode 1.1 Extended | 55.8% | Coming soon | Coming soon |
| CursorBench 3.2 | 64.9% | Coming soon | Coming soon |
| VulcanBench v3 | 87.0% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 85.9% | Coming soon | Coming soon |
| SWE-bench (Vals) | 95.4% | Coming soon | Coming soon |
| CursorBench 4.0 | 41.3% | Coming soon | Coming soon |
| SWE-Atlas Codebase QnA | Coming soon | 31.4% | Coming soon |
| SWE Multilingual | Coming soon | 77.5% | Coming soon |
| SWE-bench Verified | Coming soon | 80.7% | Coming soon |
| SciCode | Coming soon | 46.9% | Coming soon |

### Reasoning

- Winner: Not comparable
- GPT-5.6 Terra public-lane score: 69.1 (#22/28)
- Beam 1.0 501B-A23B NVFP4 public-lane score: 85.5 (Unranked · 3 rankable rows)

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 83.9% | Coming soon | Coming soon |
| ARC-AGI-3 | 0.8% | Coming soon | Coming soon |
| ARC-AGI-1 | 96.50% | Coming soon | Coming soon |
| CritPt | Coming soon | 15.4% | Coming soon |
| LongBench v2 | Coming soon | 67.2% | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- GPT-5.6 Terra public-lane score: 84.5 (#11/54)
- Beam 1.0 501B-A23B NVFP4 public-lane score: Coming soon

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 80.7% | Coming soon | Coming soon |
| MMMU-Pro w/ Python | 82% | Coming soon | Coming soon |

### Knowledge

- Winner: Not comparable
- GPT-5.6 Terra public-lane score: 70.1 (Supported · #17/184)
- Beam 1.0 501B-A23B NVFP4 public-lane score: Coming soon

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 92.9% | Coming soon | Coming soon |
| GPQA-D | 92.9% | 89.3% | GPT-5.6 Terra |
| HLE-Verified | 51.1% | Coming soon | Coming soon |
| LABBench2 | 81.2% | Coming soon | Coming soon |
| HealthBench Professional | 57.7% | Coming soon | Coming soon |
| HealthBench Hard | 32.7% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 90.9% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 86.7% | Coming soon | Coming soon |
| HLE w/o tools | Coming soon | 34.8% | Coming soon |

### Instruction Following

- Winner: Directional only
- GPT-5.6 Terra public-lane score: 87.1 (#39/148)
- Beam 1.0 501B-A23B NVFP4 public-lane score: 86.2 (#44/148)

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| IFBench | Coming soon | 79.2% | Coming soon |

### Mathematics

- Winner: Not comparable
- GPT-5.6 Terra public-lane score: 90.1 (Unranked · 3 rankable rows)
- Beam 1.0 501B-A23B NVFP4 public-lane score: 87.9 (Unranked · 1 rankable row)

| Benchmark | GPT-5.6 Terra | Beam 1.0 501B-A23B NVFP4 | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath (legacy) | 84.9% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 84.900% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 68.300% | Coming soon | Coming soon |
| AIME26 | Coming soon | 97.5% | Coming soon |

## FAQ

### Which is better overall, GPT-5.6 Terra or Beam 1.0 501B-A23B NVFP4?

GPT-5.6 Terra is ahead overall on BenchLM right now.

### Where is the biggest gap between GPT-5.6 Terra and Beam 1.0 501B-A23B NVFP4?

The widest category gap is in reasoning, where the averages are 69.1 for GPT-5.6 Terra and 85.5 for Beam 1.0 501B-A23B NVFP4.

### How many benchmarks does GPT-5.6 Terra cover on BenchLM?

GPT-5.6 Terra currently has 35 sourced benchmark scores on BenchLM.

### How many benchmarks does Beam 1.0 501B-A23B NVFP4 cover on BenchLM?

Beam 1.0 501B-A23B NVFP4 currently has 19 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.6 Terra vs GPT-5.6 Sol](/compare/gpt-5-6-sol-vs-gpt-5-6-terra)
- [Beam 1.0 501B-A23B NVFP4 vs GPT-5.6 Sol](/compare/gpt-5-6-sol-vs-reflection-beam-nvfp4)
- [GPT-5.6 Terra vs GPT-5.6 Luna](/compare/gpt-5-6-luna-vs-gpt-5-6-terra)
- [Beam 1.0 501B-A23B NVFP4 vs GPT-5.6 Luna](/compare/gpt-5-6-luna-vs-reflection-beam-nvfp4)

## Explore More

- [GPT-5.6 Terra profile](/models/gpt-5-6-terra)
- [Beam 1.0 501B-A23B NVFP4 profile](/models/reflection-beam-nvfp4)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)