Nemotron Cascade 2 30B A3B vs Mercury 2 vs MiniMax-M2.7-highspeed
Too close to call on our weighted score (Mercury 2 37, Nemotron Cascade 2 30B A3B 35, MiniMax-M2.7-highspeed 34). The right pick depends on what you value most.
NVIDIANemotron Cascade 2 30B A3B
35/100- ECI—
- Price—
- Context256K
InceptionMercury 2
37/100- ECI—
- Price$0.25 / $0.75
- Context128K
MiniMaxMiniMax-M2.7-highspeed
34/100- ECI—
- Price$0.60 / $2.40
- Context205K
Too close to call
It is close. Our weighted score puts them within 1 points (Mercury 2 37/100, Nemotron Cascade 2 30B A3B 35/100, MiniMax-M2.7-highspeed 34/100), so choose by what matters most for your work: Mercury 2 on price and Nemotron Cascade 2 30B A3B for long inputs. The score weighs inputs & features 60%, context window 40%. 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 priceMercury 2Mercury 2 $0.375 · MiniMax-M2.7-highspeed $1.05 per 1M tokens (3:1 blend) · Nemotron Cascade 2 30B A3B unpriced
- Longest contextNemotron Cascade 2 30B A3BNemotron Cascade 2 30B A3B 256,000 · MiniMax-M2.7-highspeed 204,800 · Mercury 2 128,000 tokens
- Widest inputsSame inputsNemotron Cascade 2 30B A3B: Text · Mercury 2: Text · MiniMax-M2.7-highspeed: Text
- Self-hostingNemotron Cascade 2 30B A3B and MiniMax-M2.7-highspeedPublishes downloadable weights
| Measure | Weight | Nemotron Cascade 2 30B A3B | Mercury 2 | MiniMax-M2.7-highspeed |
|---|---|---|---|---|
| Inputs & features | 60% | 35 | 45 | 35 |
| Context window | 40% | 36 | 24 | 32 |
| Overall | 100% | 35/100 | 37/100 | 34/100 |
Left out because at least one model lacks the data: capability and price. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Nemotron Cascade 2 30B A3BNVIDIA | Mercury 2Inception | MiniMax-M2.7-highspeedMiniMax |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | — | $0.25 (best) | $0.60 |
| Output | — | $0.75 (best) | $2.40 |
| Cached input | — | $0.025 (best) | $0.06 |
| Blended (3:1) | — | $0.375 (best) | $1.05 |
| Long-context rate | — | Same rate | Same rate |
| Price source | — | Official Inception API | Official MiniMax (minimax.io) API |
| Limits | |||
| Context window | 256,000 tokens (best) | 128,000 tokens | 204,800 tokens |
| Max output | 32,768 tokens | 50,000 tokens | 131,072 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yeslow · medium · high | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | Yes | No |
| Availability | |||
| Weights | Open | Proprietary | Open |
| API model ID | — | mercury-2 | MiniMax-M2.7-highspeed |
| API providers | — | 1 | 14 (best) |
| Released | Mar 24, 2026 | Feb 24, 2026 | Mar 18, 2026 |
| Knowledge cutoff | — | Jan 1, 2025 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Nemotron Cascade 2 30B A3B—
- Mercury 2$4.00
- MiniMax-M2.7-highspeed$10.80
Which should you choose?
Which is better: Nemotron Cascade 2 30B A3B, Mercury 2 or MiniMax-M2.7-highspeed?
It is close. Our weighted score puts them within 1 points (Mercury 2 37/100, Nemotron Cascade 2 30B A3B 35/100, MiniMax-M2.7-highspeed 34/100), so choose by what matters most for your work: Mercury 2 on price and Nemotron Cascade 2 30B A3B for long inputs. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Nemotron Cascade 2 30B A3B, Mercury 2 or MiniMax-M2.7-highspeed?
Mercury 2 is cheaper at $0.25 input / $0.75 output per million tokens (official Inception API price). MiniMax-M2.7-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 $0.375 per million tokens for Mercury 2 versus $1.05 for MiniMax-M2.7-highspeed (2.8× as much). Nemotron Cascade 2 30B A3B has no published per-token price.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Nemotron Cascade 2 30B A3B has not been scored yet, Mercury 2 has not been scored yet and MiniMax-M2.7-highspeed has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Nemotron Cascade 2 30B A3B, Mercury 2 and MiniMax-M2.7-highspeed 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?
Nemotron Cascade 2 30B A3B has the largest context window at 256,000 tokens, against 204,800 for MiniMax-M2.7-highspeed and 128,000 for Mercury 2. Maximum output per response: Nemotron Cascade 2 30B A3B up to 32,768, Mercury 2 up to 50,000, MiniMax-M2.7-highspeed up to 131,072 tokens.
Which can read images, PDFs, audio or video?
Nemotron Cascade 2 30B A3B accepts text; Mercury 2 accepts text; MiniMax-M2.7-highspeed accepts text. They handle the same number of input types.
Are any of these open source?
Nemotron Cascade 2 30B A3B and MiniMax-M2.7-highspeed publishes its weights and can be self-hosted; Mercury 2 is proprietary.
Which is newer?
Nemotron Cascade 2 30B A3B is the newest, released Mar 24, 2026. MiniMax-M2.7-highspeed came out Mar 18, 2026; Mercury 2 came out Feb 24, 2026. Knowledge cutoff: Mercury 2 Jan 1, 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.
