# Claude Haiku 4.5 vs Ternary Bonsai 2 27B

> Side-by-side benchmark comparison for Claude Haiku 4.5 and Ternary Bonsai 2 27B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-haiku-4-5-vs-ternary-bonsai-2-27b
- Last updated: September 18, 2026

- Shared sourced benchmarks: 1
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Evidence is too thin for a benchmark verdict: Claude Haiku 4.5 and Ternary Bonsai 2 27B share only 1 sourced benchmark row.

## Summary

- Only 1 shared sourced benchmark row is available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Ternary Bonsai 2 27B has the lower published output-token price.
- Ternary Bonsai 2 27B has the larger published context window at 262K.

## Model Snapshot

| Property | Claude Haiku 4.5 | Ternary Bonsai 2 27B |
|----------|----------|----------|
| Creator | Anthropic | Prism ML |
| Type | Proprietary | Open Weight |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 200K | 262K |
| Independent public score (not head-to-head) | 52.89 | 50.78 |
| Benchmarks Covered | 10 | 21 |
| Pricing (input/output) | $1.00 / $5.00 | $0.00 / $0.00 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: 27 (Supported · #146/154)
- Ternary Bonsai 2 27B public-lane score: 49.9 (Estimated · #59/154)

| Benchmark | Claude Haiku 4.5 | Ternary Bonsai 2 27B | Winner |
|-----------|-----------|-----------|--------|
| JobBench | 16.0% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 43.8% | Coming soon | Coming soon |
| τ²-bench results | Coming soon | 80.2% | Coming soon |
| BFCL v3 | Coming soon | 74.9% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 52.8% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: 27.4 (Supported · #147/154)
- Ternary Bonsai 2 27B public-lane score: 49.9 (Estimated · #64/154)

| Benchmark | Claude Haiku 4.5 | Ternary Bonsai 2 27B | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 73.3% | 60.8% | Claude Haiku 4.5 |
| VulcanBench v3 | 76.2% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 41.2% | Coming soon | Coming soon |
| SWE-bench (Vals) | 66.6% | Coming soon | Coming soon |
| LiveCodeBench v6 | Coming soon | 90.1% | Coming soon |
| BigCodeBench | Coming soon | 58.1% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 52.8% | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: Coming soon
- Ternary Bonsai 2 27B public-lane score: Coming soon

| Benchmark | Claude Haiku 4.5 | Ternary Bonsai 2 27B | Winner |
|-----------|-----------|-----------|--------|
| CharXiv (overall) | Coming soon | 80.0% | Coming soon |
| A-OKVQA | Coming soon | 86.8% | Coming soon |
| OmniDocBench 1.6 | Coming soon | 89.1% | Coming soon |
| RealWorldQA | Coming soon | 80.1% | Coming soon |
| OCRBench V2 | Coming soon | 56.9% | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: Coming soon
- Ternary Bonsai 2 27B public-lane score: 73.9 (Unranked · 1 rankable row)

### Knowledge

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: 44.5 (Estimated · #117/184)
- Ternary Bonsai 2 27B public-lane score: 50.6 (Estimated · #78/184)

| Benchmark | Claude Haiku 4.5 | Ternary Bonsai 2 27B | Winner |
|-----------|-----------|-----------|--------|
| GPQA Diamond (Vals) | 72.2% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 78.7% | Coming soon | Coming soon |
| MMLU-Redux | Coming soon | 89.1% | Coming soon |
| GPQA | Coming soon | 85.8% | Coming soon |
| GPQA-D | Coming soon | 85.8% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: Coming soon
- Ternary Bonsai 2 27B public-lane score: 71 (#64/124)

| Benchmark | Claude Haiku 4.5 | Ternary Bonsai 2 27B | Winner |
|-----------|-----------|-----------|--------|
| IFEval | Coming soon | 91.3% | Coming soon |
| IFBench | Coming soon | 74% | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- Claude Haiku 4.5 public-lane score: 28.8 (Unranked · 2 rankable rows)
- Ternary Bonsai 2 27B public-lane score: 76.8 (Unranked · 4 rankable rows)

| Benchmark | Claude Haiku 4.5 | Ternary Bonsai 2 27B | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 5.903% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 2.083% | Coming soon | Coming soon |
| GSM8K | Coming soon | 96.7% | Coming soon |
| MATH-500 | Coming soon | 98.8% | Coming soon |
| AIME 2025 | Coming soon | 95% | Coming soon |
| AIME26 | Coming soon | 95.8% | Coming soon |

## FAQ

### Is there enough evidence to pick between Claude Haiku 4.5 and Ternary Bonsai 2 27B?

Not yet. The models share only 1 sourced benchmark row, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [Claude Haiku 4.5 vs Claude Haiku 4.5 Thinking](/compare/claude-haiku-4-5-vs-claude-haiku-4-5-thinking)
- [Ternary Bonsai 2 27B vs Claude Haiku 4.5 Thinking](/compare/claude-haiku-4-5-thinking-vs-ternary-bonsai-2-27b)
- [Claude Haiku 4.5 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-haiku-4-5)
- [Ternary Bonsai 2 27B vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-ternary-bonsai-2-27b)

## Explore More

- [Claude Haiku 4.5 profile](/models/claude-haiku-4-5)
- [Ternary Bonsai 2 27B profile](/models/ternary-bonsai-2-27b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)