# Gemini 3.8 Flash vs MiniCPM5-2B

> Side-by-side benchmark comparison for Gemini 3.8 Flash and MiniCPM5-2B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gemini-3-8-flash-vs-minicpm5-2b
- Last updated: September 10, 2026

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

## Quick Verdict

Evidence is too thin for a benchmark verdict: Gemini 3.8 Flash and MiniCPM5-2B 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.
- MiniCPM5-2B has the lower published output-token price.
- Gemini 3.8 Flash has the larger published context window at 1M.

## Model Snapshot

| Property | Gemini 3.8 Flash | MiniCPM5-2B |
|----------|----------|----------|
| Creator | Google | OpenBMB |
| Type | Proprietary | Open Weight |
| Reasoning | Reasoning | Reasoning |
| Context | 1M | 131K |
| Independent public score (not head-to-head) | 75.61 | null |
| Benchmarks Covered | 19 | 19 |
| Pricing (input/output) | $0.75 / $3.75 | $0.00 / $0.00 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: 66.3 (Supported · #10/152)
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| Finance Agent v2 | 61.4% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 89.4% | 8.6% | Gemini 3.8 Flash |
| Terminal-Bench 4.0 | 19.10% | Coming soon | Coming soon |
| OSWorld 2.0 | 59.0% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 81.3% | Coming soon | Coming soon |
| BFCL v4 | Coming soon | 66.6% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: 66.6 (Supported · #10/151)
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| DeepSWE | 73.8% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 89.4% | 8.6% | Gemini 3.8 Flash |
| cursorBench32 | 69.2% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 89.5% | Coming soon | Coming soon |
| SWE-bench (Vals) | 80.0% | Coming soon | Coming soon |
| FrontierSWE v2 | 19.6% | Coming soon | Coming soon |
| LiveCodeBench v6 | Coming soon | 69.1% | Coming soon |
| SciCode | Coming soon | 26.3% | Coming soon |
| SWE-bench Verified | Coming soon | 46.4% | Coming soon |
| SWE-bench Pro | Coming soon | 14.4% | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: 82.8 (#8/48)
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| CharXiv w/o tools | 86.2% | Coming soon | Coming soon |
| LVBench | 87.1% | Coming soon | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: 76.9 (#8/20)
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | Coming soon | 43.7% | Coming soon |

### Knowledge

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: 74.9 (Supported · #6/183)
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| HLE-Verified | 54.9% | Coming soon | Coming soon |
| LABBench2 | 86.2% | Coming soon | Coming soon |
| BioMysteryBench (human-solvable) | 88.8% | Coming soon | Coming soon |
| BioMysteryBench (human-difficult) | 56.5% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 94.4% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 90.2% | Coming soon | Coming soon |
| MMLU-Pro | Coming soon | 70.8% | Coming soon |
| MMLU-Redux | Coming soon | 84.7% | Coming soon |
| HLE | Coming soon | 8.9% | Coming soon |
| GPQA-D | Coming soon | 70.2% | Coming soon |
| SuperGPQA | Coming soon | 40.8% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: Coming soon
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| IFBench | Coming soon | 66.3% | Coming soon |
| IFEval | Coming soon | 86.7% | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- Gemini 3.8 Flash public-lane score: Coming soon
- MiniCPM5-2B public-lane score: Coming soon

| Benchmark | Gemini 3.8 Flash | MiniCPM5-2B | Winner |
|-----------|-----------|-----------|--------|
| AIME 2025 | Coming soon | 86.5% | Coming soon |
| AIME26 | Coming soon | 86.5% | Coming soon |
| HMMT Feb 2026 | Coming soon | 63.8% | Coming soon |
| MATH-500 | Coming soon | 94.6% | Coming soon |

## FAQ

### Is there enough evidence to pick between Gemini 3.8 Flash and MiniCPM5-2B?

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

- [Gemini 3.8 Flash vs Gemini 3.8 Flash Cyber](/compare/gemini-3-8-flash-vs-gemini-3-8-flash-cyber)
- [MiniCPM5-2B vs Gemini 3.8 Flash Cyber](/compare/gemini-3-8-flash-cyber-vs-minicpm5-2b)
- [Gemini 3.8 Flash vs MiniCPM5-1B](/compare/gemini-3-8-flash-vs-minicpm5-1b)
- [MiniCPM5-2B vs MiniCPM5-1B](/compare/minicpm5-1b-vs-minicpm5-2b)

## Explore More

- [Gemini 3.8 Flash profile](/models/gemini-3-8-flash)
- [MiniCPM5-2B profile](/models/minicpm5-2b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)