# Mellum2-12B-A2.5B-Thinking vs Qwen3.5 397B

> Side-by-side benchmark comparison for Mellum2-12B-A2.5B-Thinking and Qwen3.5 397B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/mellum2-12b-a2-5b-thinking-vs-qwen3-5-397b
- Last updated: September 4, 2026

- Shared sourced benchmarks: 4
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Qwen3.5 397B if you want the stronger benchmark profile. Mellum2-12B-A2.5B-Thinking only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Qwen3.5 397B leads overall 56.44 to 52.
- The clearest category separation is in knowledge, where the averages are 48.8 for Mellum2-12B-A2.5B-Thinking and 51.6 for Qwen3.5 397B.
- The biggest single benchmark swing is GPQA in Knowledge, with scores of 57.6% and 88.4%.

## Model Snapshot

| Property | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B |
|----------|----------|----------|
| Creator | JetBrains | Alibaba |
| Type | Open Weight | Open Weight |
| Reasoning | Reasoning | Non-Reasoning |
| Context | 128K | 128K |
| Overall Score | null | 56.44 |
| Benchmarks Covered | 6 | 38 |
| Pricing (input/output) | N/A | $0.60 / $3.60 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: 48.3 (Estimated · #76/151)
- Qwen3.5 397B public-lane score: 48.2 (Estimated · #77/151)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| BFCL v4 | 45.6% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | Coming soon | 52.5% | Coming soon |
| BrowseComp | Coming soon | 62% | Coming soon |
| Claw-Eval | Coming soon | 56.8% | Coming soon |
| QwenClawBench | Coming soon | 51.8% | Coming soon |
| τ³-bench results | Coming soon | 68.4% | Coming soon |
| VITA-Bench | Coming soon | 43.7% | Coming soon |
| DeepPlanning | Coming soon | 37.6% | Coming soon |
| Toolathlon | Coming soon | 36.3% | Coming soon |
| MCP Atlas | Coming soon | 46.1% | Coming soon |
| MCP-Tasks | Coming soon | 74.2% | Coming soon |
| WideResearch | Coming soon | 74.0% | Coming soon |
| Gert Labs | Coming soon | 46.76% | Coming soon |
| ResearchClawBench | Coming soon | 14.2% | Coming soon |

### Coding

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: Coming soon
- Qwen3.5 397B public-lane score: 52.4 (Estimated · #56/183)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench v6 | 69.9% | 83.6% | Qwen3.5 397B |
| SWE-bench Verified | Coming soon | 76.2% | Coming soon |
| SWE-bench Pro | Coming soon | 50.9% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: Coming soon
- Qwen3.5 397B public-lane score: 63 (#27/48)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | Coming soon | 79% | Coming soon |
| MathVision | Coming soon | 88.6% | Coming soon |
| CharXiv | Coming soon | 80.8% | Coming soon |
| VideoMMMU | Coming soon | 84.7% | Coming soon |
| ScreenSpot Pro | Coming soon | 65.6% | Coming soon |
| V* | Coming soon | 95.8% | Coming soon |

### Reasoning

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: Coming soon
- Qwen3.5 397B public-lane score: 59.8 (Unranked · 2 rankable rows)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | Coming soon | 63.2% | Coming soon |
| AI-Needle | Coming soon | 68.7% | Coming soon |

### Knowledge

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: 48.8 (Estimated · #96/181)
- Qwen3.5 397B public-lane score: 51.6 (Estimated · #80/181)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| MMLU-Redux | 86.2% | 94.9% | Qwen3.5 397B |
| GPQA | 57.6% | 88.4% | Qwen3.5 397B |
| GPQA-D | 57.6% | Coming soon | Coming soon |
| SuperGPQA | Coming soon | 70.4% | Coming soon |
| MMLU-Pro | Coming soon | 87.8% | Coming soon |
| C-Eval | Coming soon | 93% | Coming soon |
| HLE | Coming soon | 28.7% | Coming soon |

### Instruction Following

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: Coming soon
- Qwen3.5 397B public-lane score: 0 (#120/120)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 76.5% | 92.6% | Qwen3.5 397B |

### Multilingual

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: Coming soon
- Qwen3.5 397B public-lane score: 69.7 (#5/12)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| MMLU-ProX | Coming soon | 84.7% | Coming soon |
| NOVA-63 | Coming soon | 59.1% | Coming soon |

### Mathematics

- Winner: Coming soon
- Mellum2-12B-A2.5B-Thinking public-lane score: Coming soon
- Qwen3.5 397B public-lane score: 74.2 (Unranked · 5 rankable rows)

| Benchmark | Mellum2-12B-A2.5B-Thinking | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | Coming soon | 93.3% | Coming soon |
| HMMT Feb 2025 | Coming soon | 94.8% | Coming soon |
| HMMT Nov 2025 | Coming soon | 92.7% | Coming soon |
| HMMT Feb 2026 | Coming soon | 87.9% | Coming soon |
| MMAnswerBench | Coming soon | 80.9% | Coming soon |

## FAQ

### Which is better overall, Mellum2-12B-A2.5B-Thinking or Qwen3.5 397B?

Qwen3.5 397B is ahead overall on BenchLM right now.

### Where is the biggest gap between Mellum2-12B-A2.5B-Thinking and Qwen3.5 397B?

The widest category gap is in knowledge, where the averages are 48.8 for Mellum2-12B-A2.5B-Thinking and 51.6 for Qwen3.5 397B.

### How many benchmarks does Mellum2-12B-A2.5B-Thinking cover on BenchLM?

Mellum2-12B-A2.5B-Thinking currently has 6 sourced benchmark scores on BenchLM.

### How many benchmarks does Qwen3.5 397B cover on BenchLM?

Qwen3.5 397B currently has 38 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Mellum2-12B-A2.5B-Thinking vs Mellum2-12B-A2.5B-Instruct](/compare/mellum2-12b-a2-5b-instruct-vs-mellum2-12b-a2-5b-thinking)
- [Qwen3.5 397B vs Mellum2-12B-A2.5B-Instruct](/compare/mellum2-12b-a2-5b-instruct-vs-qwen3-5-397b)
- [Mellum2-12B-A2.5B-Thinking vs Qwen3.5 397B (Reasoning)](/compare/mellum2-12b-a2-5b-thinking-vs-qwen3-5-397b-reasoning)
- [Qwen3.5 397B vs Qwen3.5 397B (Reasoning)](/compare/qwen3-5-397b-vs-qwen3-5-397b-reasoning)

## Explore More

- [Mellum2-12B-A2.5B-Thinking profile](/models/mellum2-12b-a2-5b-thinking)
- [Qwen3.5 397B profile](/models/qwen3-5-397b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)