# Mercury 3 vs Beam 1.0 501B-A23B

> Side-by-side benchmark comparison for Mercury 3 and Beam 1.0 501B-A23B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/mercury-3-vs-reflection-beam
- Last updated: October 19, 2026

- Shared sourced benchmarks: 5
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Beam 1.0 501B-A23B leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- Beam 1.0 501B-A23B has the higher overall point estimate 51.58 to 0. Conditional score ranges do not establish rank confidence.
- The biggest single benchmark swing is DeepSearchQA in Agentic, with scores of 71.3% and 80.5%.
- Beam 1.0 501B-A23B is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | Mercury 3 | Beam 1.0 501B-A23B |
|----------|----------|----------|
| Creator | Inception | Reflection AI |
| Type | Proprietary | Open Weight |
| Reasoning | Reasoning | Reasoning |
| Context | 1M | 1M |
| Overall Score | null | 51.58 |
| Benchmarks Covered | 8 | 19 |
| Pricing (input/output) | $0.10 / $0.35 | Pricing unavailable / Pricing unavailable |

## Category Breakdown

### Agentic

- Winner: Not comparable
- Mercury 3 public-lane score: Coming soon
- Beam 1.0 501B-A23B public-lane score: 49.9 (Supported · #49/130)

| Benchmark | Mercury 3 | Beam 1.0 501B-A23B | Winner |
|-----------|-----------|-----------|--------|
| τ³-bench results | 30.2% | 36.1% | Beam 1.0 501B-A23B |
| BrowseComp+ | 55.18% | Coming soon | Coming soon |
| DeepSearchQA | 71.3% | 80.5% | Beam 1.0 501B-A23B |
| HLE w/ tools | Coming soon | 51.1% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 80.1% | Coming soon |
| MCP Atlas | Coming soon | 80.8% | Coming soon |
| BrowseComp | Coming soon | 76.8% | Coming soon |

### Coding

- Winner: Not comparable
- Mercury 3 public-lane score: Coming soon
- Beam 1.0 501B-A23B public-lane score: Coming soon

| Benchmark | Mercury 3 | Beam 1.0 501B-A23B | Winner |
|-----------|-----------|-----------|--------|
| DeepSWE | 46.3% | 43.3% | Mercury 3 |
| SWE-bench Pro | Coming soon | 66.1% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 80.1% | Coming soon |
| SWE-Atlas Codebase QnA | Coming soon | 31.8% | Coming soon |
| SWE Multilingual | Coming soon | 78.3% | Coming soon |
| SWE-bench Verified | Coming soon | 81.6% | Coming soon |
| SciCode | Coming soon | 49.2% | Coming soon |

### Reasoning

- Winner: Not comparable
- Mercury 3 public-lane score: Coming soon
- Beam 1.0 501B-A23B public-lane score: 78.7 (Unranked · 2 rankable rows)

| Benchmark | Mercury 3 | Beam 1.0 501B-A23B | Winner |
|-----------|-----------|-----------|--------|
| MRCRv2 | 75.5% | Coming soon | Coming soon |
| CritPt | Coming soon | 16.3% | Coming soon |
| LongBench v2 | Coming soon | 67.5% | Coming soon |

### Knowledge

- Winner: Not comparable
- Mercury 3 public-lane score: Coming soon
- Beam 1.0 501B-A23B public-lane score: 55.5 (Supported · #59/186)

| Benchmark | Mercury 3 | Beam 1.0 501B-A23B | Winner |
|-----------|-----------|-----------|--------|
| GPQA-D | 89.2% | 90.6% | Beam 1.0 501B-A23B |
| HLE w/o tools | Coming soon | 36% | Coming soon |

### Instruction Following

- Winner: Not comparable
- Mercury 3 public-lane score: Coming soon
- Beam 1.0 501B-A23B public-lane score: Coming soon

| Benchmark | Mercury 3 | Beam 1.0 501B-A23B | Winner |
|-----------|-----------|-----------|--------|
| IFBench | 76.8% | 79.6% | Beam 1.0 501B-A23B |
| MultiChallenge (provider reports) | 60.35% | Coming soon | Coming soon |

### Mathematics

- Winner: Not comparable
- Mercury 3 public-lane score: Coming soon
- Beam 1.0 501B-A23B public-lane score: Coming soon

| Benchmark | Mercury 3 | Beam 1.0 501B-A23B | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | Coming soon | 97.7% | Coming soon |

## FAQ

### Which is better overall, Mercury 3 or Beam 1.0 501B-A23B?

Beam 1.0 501B-A23B has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### How many benchmarks does Mercury 3 cover on BenchLM?

Mercury 3 currently has 8 sourced benchmark scores on BenchLM.

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

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

## Related Comparisons

- [Mercury 3 vs Beam 1.0 501B-A23B FP8](/compare/mercury-3-vs-reflection-beam-fp8)
- [Beam 1.0 501B-A23B vs Beam 1.0 501B-A23B FP8](/compare/reflection-beam-vs-reflection-beam-fp8)
- [Mercury 3 vs Beam 1.0 501B-A23B NVFP4](/compare/mercury-3-vs-reflection-beam-nvfp4)
- [Beam 1.0 501B-A23B vs Beam 1.0 501B-A23B NVFP4](/compare/reflection-beam-vs-reflection-beam-nvfp4)

## Explore More

- [Mercury 3 profile](/models/mercury-3)
- [Beam 1.0 501B-A23B profile](/models/reflection-beam)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)