# Beam vs Sakana Fugu-Ultra

> Side-by-side benchmark comparison for Beam and Sakana Fugu-Ultra across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/reflection-beam-vs-sakana-fugu-ultra
- Last updated: October 5, 2026

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

## Quick Verdict

Pick Sakana Fugu-Ultra if you want the stronger benchmark profile. Beam only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Sakana Fugu-Ultra leads overall 75 to 0.
- The biggest single benchmark swing is HLE w/o tools in Knowledge, with scores of 36.2% and 50%.
- Sakana Fugu-Ultra also has the larger context window at 1M.

## Model Snapshot

| Property | Beam | Sakana Fugu-Ultra |
|----------|----------|----------|
| Creator | Reflection AI | Sakana AI |
| Type | Pending | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | null | 1M |
| Overall Score | null | null |
| Benchmarks Covered | 16 | 11 |
| Pricing (input/output) | Pricing unavailable / Pricing unavailable | Pricing unavailable / Pricing unavailable |

## Category Breakdown

### Agentic

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.1 | 80.1% | 82.1% | Sakana Fugu-Ultra |
| MCP Atlas | 78.7% | Coming soon | Coming soon |
| BrowseComp | 77.4% | Coming soon | Coming soon |
| DeepSearchQA | 80.1% | Coming soon | Coming soon |

### Coding

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| DeepSWE | 44.4% | Coming soon | Coming soon |
| SWE-bench Pro | 65.5% | 73.7% | Sakana Fugu-Ultra |
| Terminal-Bench 2.1 | 80.1% | 82.1% | Sakana Fugu-Ultra |
| SWE-Atlas Codebase QnA | 34.6% | Coming soon | Coming soon |
| SWE Multilingual | 78% | Coming soon | Coming soon |
| SWE-bench Verified | 80.9% | Coming soon | Coming soon |
| SciCode | 49.7% | 58.7% | Sakana Fugu-Ultra |
| LiveCodeBench v6 | Coming soon | 93.2% | Coming soon |
| LiveCodeBench Pro | Coming soon | 90.8% | Coming soon |

### Reasoning

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: 79 (Unranked · 1 rankable row)

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | 65.5% | Coming soon | Coming soon |
| MRCRv2 | Coming soon | 93.6% | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: 73.6 (Unranked · 1 rankable row)

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| CharXiv | Coming soon | 86.6% | Coming soon |

### Knowledge

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| GPQA-D | 90.5% | 95.5% | Sakana Fugu-Ultra |
| HLE w/o tools | 36.2% | 50% | Sakana Fugu-Ultra |
| GPQA | Coming soon | 95.5% | Coming soon |

### Instruction Following

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| IFBench | 79.7% | Coming soon | Coming soon |

### Mathematics

- Winner: Not comparable
- Beam public-lane score: Coming soon
- Sakana Fugu-Ultra public-lane score: Coming soon

| Benchmark | Beam | Sakana Fugu-Ultra | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | 97.8% | Coming soon | Coming soon |

## FAQ

### Which is better overall, Beam or Sakana Fugu-Ultra?

Sakana Fugu-Ultra is ahead overall on BenchLM right now.

### How many benchmarks does Beam cover on BenchLM?

Beam currently has 16 sourced benchmark scores on BenchLM.

### How many benchmarks does Sakana Fugu-Ultra cover on BenchLM?

Sakana Fugu-Ultra currently has 11 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Beam vs Sakana Fugu](/compare/reflection-beam-vs-sakana-fugu)
- [Sakana Fugu-Ultra vs Sakana Fugu](/compare/sakana-fugu-vs-sakana-fugu-ultra)
- [Beam vs Fugu Cyber](/compare/reflection-beam-vs-sakana-fugu-cyber)
- [Sakana Fugu-Ultra vs Fugu Cyber](/compare/sakana-fugu-cyber-vs-sakana-fugu-ultra)

## Explore More

- [Beam profile](/models/reflection-beam)
- [Sakana Fugu-Ultra profile](/models/sakana-fugu-ultra)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)