DeepSeek OCR 2 vs MiniMax-M2.5-highspeed vs Voxtral Mini 3B 2507
Voxtral Mini 3B 2507 comes out ahead, 61 to 41 and 32 on our weighted score, and it is the cheaper option too.
DeepSeekDeepSeek OCR 2
32/100- ECI—
- Price$0.89 / $1.47
- Context8K
MiniMaxMiniMax-M2.5-highspeed
41/100- ECI—
- Price$0.60 / $2.40
- Context205K
- Our pick
Mistral AIVoxtral Mini 3B 2507
61/100- ECI—
- Price$0.04 / $0.04
- Context33K
Voxtral Mini 3B 2507 is our pick
Voxtral Mini 3B 2507 is the better all-round choice, scoring 61/100 against MiniMax-M2.5-highspeed (41) and DeepSeek OCR 2 (32). It leads on price. MiniMax-M2.5-highspeed wins on context window. 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 priceVoxtral Mini 3B 2507Voxtral Mini 3B 2507 $0.04 · DeepSeek OCR 2 $1.03 · MiniMax-M2.5-highspeed $1.05 per 1M tokens (3:1 blend)
- Longest contextMiniMax-M2.5-highspeedMiniMax-M2.5-highspeed 204,800 · Voxtral Mini 3B 2507 32,768 · DeepSeek OCR 2 8,192 tokens
- Widest inputsDeepSeek OCR 2 and Voxtral Mini 3B 2507DeepSeek OCR 2: Text, Images · MiniMax-M2.5-highspeed: Text · Voxtral Mini 3B 2507: Text, Audio
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | DeepSeek OCR 2 | MiniMax-M2.5-highspeed | Voxtral Mini 3B 2507 |
|---|---|---|---|---|
| Price | 50% | 49 | 49 | 100 |
| Inputs & features | 30% | 25 | 35 | 35 |
| Context window | 20% | 0 | 32 | 0 |
| Overall | 100% | 32/100 | 41/100 | 61/100 |
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 | DeepSeek OCR 2DeepSeek | MiniMax-M2.5-highspeedMiniMax | Voxtral Mini 3B 2507Mistral AI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.89 | $0.60 | $0.04 (best) |
| Output | $1.47 | $2.40 | $0.04 (best) |
| Cached input | — | $0.06 | — |
| Blended (3:1) | $1.03 | $1.05 | $0.04 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 1 providers | Official MiniMax (minimax.io) API | Median of 2 providers |
| Limits | |||
| Context window | 8,192 tokens | 204,800 tokens (best) | 32,768 tokens |
| Max output | 8,192 tokens | 131,072 tokens (best) | 32,768 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | No |
| PDFs | No | No | No |
| Audio | No | No | Yes |
| Video | No | No | No |
| Reasoning | No | Yes | No |
| Tool calling | No | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Open | OpenApache 2.0 |
| API model ID | — | MiniMax-M2.5-highspeed | — |
| API providers | 2 | 7 (best) | 2 |
| Released | Jan 27, 2026 | Feb 13, 2026 | Jul 15, 2025 |
| 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.
- DeepSeek OCR 2$11.83
- MiniMax-M2.5-highspeed$10.80
- Voxtral Mini 3B 2507$0.48
Which should you choose?
Which is better: DeepSeek OCR 2, MiniMax-M2.5-highspeed or Voxtral Mini 3B 2507?
Voxtral Mini 3B 2507 is the better all-round choice, scoring 61/100 against MiniMax-M2.5-highspeed (41) and DeepSeek OCR 2 (32). It leads on price. MiniMax-M2.5-highspeed wins on context window. 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, DeepSeek OCR 2, MiniMax-M2.5-highspeed or Voxtral Mini 3B 2507?
Voxtral Mini 3B 2507 is cheaper at $0.04 input / $0.04 output per million tokens (median across 2 API providers). DeepSeek OCR 2 costs $0.89 input / $1.47 output per million tokens (median across 1 API provider); MiniMax-M2.5-highspeed costs $0.60 input / $2.40 output per million tokens (official MiniMax (minimax.io) API price). At a typical mix of three input tokens to one output token, that is $0.04 per million tokens for Voxtral Mini 3B 2507 versus $1.03 for DeepSeek OCR 2 (26× as much) and $1.05 for MiniMax-M2.5-highspeed (26× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. DeepSeek OCR 2 has not been scored yet, MiniMax-M2.5-highspeed has not been scored yet and Voxtral Mini 3B 2507 has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for DeepSeek OCR 2, MiniMax-M2.5-highspeed and Voxtral Mini 3B 2507 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that DeepSeek OCR 2 does not support tool calling, which most coding agents need.
Which has the bigger context window?
MiniMax-M2.5-highspeed has the largest context window at 204,800 tokens, against 32,768 for Voxtral Mini 3B 2507 and 8,192 for DeepSeek OCR 2. Maximum output per response: DeepSeek OCR 2 up to 8,192, MiniMax-M2.5-highspeed up to 131,072, Voxtral Mini 3B 2507 up to 32,768 tokens.
Which can read images, PDFs, audio or video?
DeepSeek OCR 2 accepts text and images; MiniMax-M2.5-highspeed accepts text; Voxtral Mini 3B 2507 accepts text and audio. DeepSeek OCR 2 handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights (Apache 2.0), so you can self-host them.
Which is newer?
MiniMax-M2.5-highspeed is the newest, released Feb 13, 2026. DeepSeek OCR 2 came out Jan 27, 2026; Voxtral Mini 3B 2507 came out Jul 15, 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.
