MiMo-V2.5 vs Step 3.5 Flash 2603 vs Trinity Large Thinking
MiMo-V2.5 comes out ahead, 79 to 62 and 55 on our weighted score, though Step 3.5 Flash 2603 is 14% cheaper per token.
- Our pick
XiaomiMiMo-V2.5
79/100- ECI—
- Price$0.14 / $0.28
- Context1.05M
StepFunStep 3.5 Flash 2603
62/100- ECI—
- Price$0.10 / $0.30
- Context256K
Arcee AITrinity Large Thinking
55/100- ECI—
- Price$0.25 / $0.80
- Context524K
MiMo-V2.5 is our pick
MiMo-V2.5 is the better all-round choice, scoring 79/100 against Step 3.5 Flash 2603 (62) and Trinity Large Thinking (55). It leads on inputs & features and context window. Step 3.5 Flash 2603 wins on price. 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 priceStep 3.5 Flash 2603Step 3.5 Flash 2603 $0.15 · MiMo-V2.5 $0.175 · Trinity Large Thinking $0.388 per 1M tokens (3:1 blend)
- Longest contextMiMo-V2.5MiMo-V2.5 1,048,576 · Trinity Large Thinking 524,288 · Step 3.5 Flash 2603 256,000 tokens
- Widest inputsMiMo-V2.5MiMo-V2.5: Text, Images, Audio, Video · Step 3.5 Flash 2603: Text · Trinity Large Thinking: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | MiMo-V2.5 | Step 3.5 Flash 2603 | Trinity Large Thinking |
|---|---|---|---|---|
| Price | 50% | 86 | 89 | 69 |
| Inputs & features | 30% | 80 | 35 | 35 |
| Context window | 20% | 61 | 36 | 49 |
| Overall | 100% | 79/100 | 62/100 | 55/100 |
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 | MiMo-V2.5Xiaomi | Step 3.5 Flash 2603StepFun | Trinity Large ThinkingArcee AI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.14 | $0.10 (best) | $0.25 |
| Output | $0.28 (best) | $0.30 | $0.80 |
| Cached input | $0.0028 (best) | $0.02 | $0.06 |
| Blended (3:1) | $0.175 | $0.15 (best) | $0.388 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Xiaomi API | Official StepFun (Global) API | Official Arcee API |
| Limits | |||
| Context window | 1,048,576 tokens (best) | 256,000 tokens | 524,288 tokens |
| Max output | 131,072 tokens | 256,000 tokens | 262,144 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | No |
| PDFs | No | No | No |
| Audio | Yes | No | No |
| Video | Yes | No | No |
| Reasoning | Yes | Yeslow · high | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Open | OpenOpenMDW-1.1 |
| API model ID | mimo-v2.5 | step-3.5-flash-2603 | trinity-large-thinking |
| API providers | 21 (best) | 3 | 6 |
| Released | Apr 22, 2026 | Apr 2, 2026 | Apr 1, 2026 |
| Knowledge cutoff | Dec 2024 | Jan 2025 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- MiMo-V2.5$1.96
- Step 3.5 Flash 2603$1.60
- Trinity Large Thinking$4.10
Which should you choose?
Which is better: MiMo-V2.5, Step 3.5 Flash 2603 or Trinity Large Thinking?
MiMo-V2.5 is the better all-round choice, scoring 79/100 against Step 3.5 Flash 2603 (62) and Trinity Large Thinking (55). It leads on inputs & features and context window. Step 3.5 Flash 2603 wins on price. 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, MiMo-V2.5, Step 3.5 Flash 2603 or Trinity Large Thinking?
Step 3.5 Flash 2603 is cheaper at $0.10 input / $0.30 output per million tokens (official StepFun (Global) API price). MiMo-V2.5 costs $0.14 input / $0.28 output per million tokens (official Xiaomi API price); Trinity Large Thinking costs $0.25 input / $0.80 output per million tokens (official Arcee API price). At a typical mix of three input tokens to one output token, that is $0.15 per million tokens for Step 3.5 Flash 2603 versus $0.175 for MiMo-V2.5 (1.2× as much) and $0.388 for Trinity Large Thinking (2.6× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. MiMo-V2.5 has not been scored yet, Step 3.5 Flash 2603 has not been scored yet and Trinity Large Thinking has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for MiMo-V2.5, Step 3.5 Flash 2603 and Trinity Large Thinking 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?
MiMo-V2.5 has the largest context window at 1,048,576 tokens, against 524,288 for Trinity Large Thinking and 256,000 for Step 3.5 Flash 2603. Maximum output per response: MiMo-V2.5 up to 131,072, Step 3.5 Flash 2603 up to 256,000, Trinity Large Thinking up to 262,144 tokens.
Which can read images, PDFs, audio or video?
MiMo-V2.5 accepts text, images, audio and video; Step 3.5 Flash 2603 accepts text; Trinity Large Thinking accepts text. MiMo-V2.5 handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights (OpenMDW-1.1), so you can self-host them.
Which is newer?
MiMo-V2.5 is the newest, released Apr 22, 2026. Step 3.5 Flash 2603 came out Apr 2, 2026; Trinity Large Thinking came out Apr 1, 2026. Knowledge cutoff: MiMo-V2.5 Dec 2024, Step 3.5 Flash 2603 Jan 2025.
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.
