MiniCPM5-2B vs Qwen3.8 Flash Next vs DeepSeek V4 Flash Vision Exp
Too close to call on our weighted score (DeepSeek V4 Flash Vision Exp 70, Qwen3.8 Flash Next 70, MiniCPM5-2B 53). The right pick depends on what you value most.
OpenBMB
MiniCPM5-2B
53/100- ECI—
- Price$0.124 / $0.743
- Context131K
Alibaba (Qwen)Qwen3.8 Flash Next
70/100- ECI—
- Price$0.20 / $0.50
- Context262K
DeepSeekDeepSeek V4 Flash Vision Exp
70/100- ECI—
- Price$0.216 / $0.647
- Context1M
Too close to call
It is close. Our weighted score puts them within a point (DeepSeek V4 Flash Vision Exp 70/100, Qwen3.8 Flash Next 70/100, MiniCPM5-2B 53/100), so choose by what matters most for your work: Qwen3.8 Flash Next on price and DeepSeek V4 Flash Vision Exp for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceQwen3.8 Flash NextQwen3.8 Flash Next $0.275 · MiniCPM5-2B $0.279 · DeepSeek V4 Flash Vision Exp $0.323 per 1M tokens (3:1 blend)
- Longest contextDeepSeek V4 Flash Vision ExpDeepSeek V4 Flash Vision Exp 1,000,000 · Qwen3.8 Flash Next 262,144 · MiniCPM5-2B 131,072 tokens
- Widest inputsQwen3.8 Flash NextMiniCPM5-2B: Text · Qwen3.8 Flash Next: Text, Images, Video · DeepSeek V4 Flash Vision Exp: Text, Images
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | MiniCPM5-2B | Qwen3.8 Flash Next | DeepSeek V4 Flash Vision Exp |
|---|---|---|---|---|
| Price | 50% | 76 | 76 | 73 |
| Inputs & features | 30% | 35 | 80 | 70 |
| Context window | 20% | 24 | 37 | 60 |
| Overall | 100% | 53/100 | 70/100 | 70/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | MiniCPM5-2BOpenBMB | Qwen3.8 Flash NextAlibaba (Qwen) | DeepSeek V4 Flash Vision ExpDeepSeek |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.124 (best) | $0.20 | $0.216 |
| Output | $0.743 | $0.50 (best) | $0.647 |
| Cached input | — | — | — |
| Blended (3:1) | $0.279 | $0.275 (best) | $0.323 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 1 providers | Median of 5 providers | Median of 15 providers |
| Limits | |||
| Context window | 131,072 tokens | 262,144 tokens | 1,000,000 tokens (best) |
| Max output | 131,072 tokens | 131,072 tokens | 384,000 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | Yes | Yes | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | Yes | Yes |
| Availability | |||
| Weights | Openapache-2.0 | Openqwen-community-1.0 | OpenMIT |
| API model ID | — | — | — |
| API providers | 1 | 5 | 15 (best) |
| Released | Sep 6, 2026 | Aug 27, 2026 | Aug 21, 2026 |
| Knowledge cutoff | — | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- MiniCPM5-2B$2.73
- Qwen3.8 Flash Next$3.00
- DeepSeek V4 Flash Vision Exp$3.45
Which should you choose?
Which is better: MiniCPM5-2B, Qwen3.8 Flash Next or DeepSeek V4 Flash Vision Exp?
It is close. Our weighted score puts them within a point (DeepSeek V4 Flash Vision Exp 70/100, Qwen3.8 Flash Next 70/100, MiniCPM5-2B 53/100), so choose by what matters most for your work: Qwen3.8 Flash Next on price and DeepSeek V4 Flash Vision Exp for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, MiniCPM5-2B, Qwen3.8 Flash Next or DeepSeek V4 Flash Vision Exp?
Qwen3.8 Flash Next is cheaper at $0.20 input / $0.50 output per million tokens (median across 5 API providers). MiniCPM5-2B costs $0.124 input / $0.743 output per million tokens (median across 1 API provider); DeepSeek V4 Flash Vision Exp costs $0.216 input / $0.647 output per million tokens (median across 15 API providers). At a typical mix of three input tokens to one output token, that is $0.275 per million tokens for Qwen3.8 Flash Next versus $0.279 for MiniCPM5-2B (1× as much) and $0.323 for DeepSeek V4 Flash Vision Exp (1.2× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. MiniCPM5-2B has not been scored yet, Qwen3.8 Flash Next has not been scored yet and DeepSeek V4 Flash Vision Exp has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for MiniCPM5-2B, Qwen3.8 Flash Next and DeepSeek V4 Flash Vision Exp yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
DeepSeek V4 Flash Vision Exp has the largest context window at 1,000,000 tokens, against 262,144 for Qwen3.8 Flash Next and 131,072 for MiniCPM5-2B. Maximum output per response: MiniCPM5-2B up to 131,072, Qwen3.8 Flash Next up to 131,072, DeepSeek V4 Flash Vision Exp up to 384,000 tokens.
Which can read images, PDFs, audio or video?
MiniCPM5-2B accepts text; Qwen3.8 Flash Next accepts text, images and video; DeepSeek V4 Flash Vision Exp accepts text and images. Qwen3.8 Flash Next handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights (apache-2.0, qwen-community-1.0 and MIT), so you can self-host them.
Which is newer?
MiniCPM5-2B is the newest, released Sep 6, 2026. Qwen3.8 Flash Next came out Aug 27, 2026; DeepSeek V4 Flash Vision Exp came out Aug 21, 2026.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
