# Hy3 Preview vs Mellum2-12B-A2.5B-Instruct

> Side-by-side benchmark comparison for Hy3 Preview and Mellum2-12B-A2.5B-Instruct across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/hy3-preview-vs-mellum2-12b-a2-5b-instruct
- Last updated: September 4, 2026

- Shared sourced benchmarks: 2
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Evidence is too thin for a benchmark verdict: Hy3 Preview and Mellum2-12B-A2.5B-Instruct share only 2 sourced benchmark rows.

## Summary

- Only 2 shared sourced benchmark rows are available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Hy3 Preview has the larger published context window at 256K.

## Model Snapshot

| Property | Hy3 Preview | Mellum2-12B-A2.5B-Instruct |
|----------|----------|----------|
| Creator | Tencent | JetBrains |
| Type | Open Weight | Open Weight |
| Reasoning | Reasoning | Non-Reasoning |
| Context | 256K | 128K |
| Independent public score (not head-to-head) | 57.08 | null |
| Benchmarks Covered | 6 | 6 |
| Pricing (input/output) | $0.00 / $0.00 | N/A |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Hy3 Preview public-lane score: 49.9 (Estimated · #66/151)
- Mellum2-12B-A2.5B-Instruct public-lane score: 47.8 (Estimated · #82/151)

| Benchmark | Hy3 Preview | Mellum2-12B-A2.5B-Instruct | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 54.4% | Coming soon | Coming soon |
| Gert Labs | 36.91% | Coming soon | Coming soon |
| BFCL v4 | Coming soon | 44.2% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Hy3 Preview public-lane score: 45.9 (Estimated · #103/183)
- Mellum2-12B-A2.5B-Instruct public-lane score: Coming soon

| Benchmark | Hy3 Preview | Mellum2-12B-A2.5B-Instruct | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 74.4% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 54.4% | Coming soon | Coming soon |
| LiveCodeBench v6 | Coming soon | 37.2% | Coming soon |

### Knowledge

- Winner: Insufficient shared evidence
- Hy3 Preview public-lane score: 50.6 (Estimated · #86/181)
- Mellum2-12B-A2.5B-Instruct public-lane score: 48.1 (Estimated · #103/181)

| Benchmark | Hy3 Preview | Mellum2-12B-A2.5B-Instruct | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 87.2% | 40.9% | Hy3 Preview |
| GPQA-D | 87.2% | 40.9% | Hy3 Preview |
| MMLU-Redux | Coming soon | 78.1% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- Hy3 Preview public-lane score: 54.8 (#74/120)
- Mellum2-12B-A2.5B-Instruct public-lane score: Coming soon

| Benchmark | Hy3 Preview | Mellum2-12B-A2.5B-Instruct | Winner |
|-----------|-----------|-----------|--------|
| IFEval | Coming soon | 75.8% | Coming soon |

## FAQ

### Is there enough evidence to pick between Hy3 Preview and Mellum2-12B-A2.5B-Instruct?

Not yet. The models share only 2 sourced benchmark rows, 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

- [Hy3 Preview vs Hy3](/compare/hy3-vs-hy3-preview)
- [Mellum2-12B-A2.5B-Instruct vs Hy3](/compare/hy3-vs-mellum2-12b-a2-5b-instruct)
- [Hy3 Preview vs Mellum2-12B-A2.5B-Thinking](/compare/hy3-preview-vs-mellum2-12b-a2-5b-thinking)
- [Mellum2-12B-A2.5B-Instruct vs Mellum2-12B-A2.5B-Thinking](/compare/mellum2-12b-a2-5b-instruct-vs-mellum2-12b-a2-5b-thinking)

## Explore More

- [Hy3 Preview profile](/models/hy3-preview)
- [Mellum2-12B-A2.5B-Instruct profile](/models/mellum2-12b-a2-5b-instruct)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)