Seed 2.0 Mini vs Nemotron 3 Super 120B A12B vs Ministral 3 3B
Seed 2.0 Mini comes out ahead, 80 to 74 and 54 on our weighted score, and it is the cheaper option too.
- Our pick
ByteDance SeedSeed 2.0 Mini
80/100- ECI—
- Price$0.03 / $0.297
- Context256K
NVIDIANemotron 3 Super 120B A12B
54/100- ECI—
- Price$0.20 / $0.80
- Context262K
Mistral AIMinistral 3 3B
74/100- ECI—
- Price$0.10 / $0.10
- Context262K
Seed 2.0 Mini is our pick
Seed 2.0 Mini is the better all-round choice, scoring 80/100 against Ministral 3 3B (74) and Nemotron 3 Super 120B A12B (54). It leads on inputs & features. 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 priceSeed 2.0 MiniSeed 2.0 Mini $0.097 · Ministral 3 3B $0.10 · Nemotron 3 Super 120B A12B $0.35 per 1M tokens (3:1 blend)
- Longest contextNemotron 3 Super 120B A12B and Ministral 3 3BNemotron 3 Super 120B A12B 262,144 · Ministral 3 3B 262,144 · Seed 2.0 Mini 256,000 tokens
- Widest inputsSeed 2.0 MiniSeed 2.0 Mini: Text, Images, Video · Nemotron 3 Super 120B A12B: Text · Ministral 3 3B: Text, Images
- Self-hostingNemotron 3 Super 120B A12B and Ministral 3 3BPublishes downloadable weights (Apache 2.0)
| Measure | Weight | Seed 2.0 Mini | Nemotron 3 Super 120B A12B | Ministral 3 3B |
|---|---|---|---|---|
| Price | 50% | 98 | 72 | 97 |
| Inputs & features | 30% | 80 | 35 | 60 |
| Context window | 20% | 36 | 37 | 37 |
| Overall | 100% | 80/100 | 54/100 | 74/100 |
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 | Seed 2.0 MiniByteDance Seed | Nemotron 3 Super 120B A12BNVIDIA | Ministral 3 3BMistral AI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.03 (best) | $0.20 | $0.10 |
| Output | $0.297 | $0.80 | $0.10 (best) |
| Cached input | $0.0059 | — | — |
| Blended (3:1) | $0.097 (best) | $0.35 | $0.10 |
| Long-context rate | Over 32K: $0.059 / $0.594 | Same rate | Same rate |
| Price source | Official Volcengine Ark API | Official Nvidia API | Median of 1 providers |
| Limits | |||
| Context window | 256,000 tokens | 262,144 tokens (best) | 262,144 tokens (best) |
| Max output | 32,000 tokens | 262,144 tokens (best) | 262,144 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | Yes | No | No |
| Reasoning | Yesminimal · low · medium · high · xhigh · max | Yes | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | No | Yes |
| Availability | |||
| Weights | Proprietary | Open | OpenApache 2.0 |
| API model ID | doubao-seed-2-0-mini-260428 | nvidia/nemotron-3-super-120b-a12b | — |
| API providers | 8 | 19 (best) | 1 |
| Released | Feb 14, 2026 | Mar 11, 2026 | Dec 2, 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.
- Seed 2.0 Mini$0.891
- Nemotron 3 Super 120B A12B$3.60
- Ministral 3 3B$1.20
Which should you choose?
Which is better: Seed 2.0 Mini, Nemotron 3 Super 120B A12B or Ministral 3 3B?
Seed 2.0 Mini is the better all-round choice, scoring 80/100 against Ministral 3 3B (74) and Nemotron 3 Super 120B A12B (54). It leads on inputs & features. 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, Seed 2.0 Mini, Nemotron 3 Super 120B A12B or Ministral 3 3B?
Seed 2.0 Mini is cheaper at $0.03 input / $0.297 output per million tokens (official Volcengine Ark API price). Ministral 3 3B costs $0.10 input / $0.10 output per million tokens (median across 1 API provider); Nemotron 3 Super 120B A12B costs $0.20 input / $0.80 output per million tokens (official Nvidia API price). At a typical mix of three input tokens to one output token, that is $0.097 per million tokens for Seed 2.0 Mini versus $0.10 for Ministral 3 3B (1× as much) and $0.35 for Nemotron 3 Super 120B A12B (3.6× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Seed 2.0 Mini has not been scored yet, Nemotron 3 Super 120B A12B has not been scored yet and Ministral 3 3B has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Seed 2.0 Mini, Nemotron 3 Super 120B A12B and Ministral 3 3B 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 3 Super 120B A12B and Ministral 3 3B have the largest context windows (262,144 and 262,144 tokens), against 256,000 for Seed 2.0 Mini. Maximum output per response: Seed 2.0 Mini up to 32,000, Nemotron 3 Super 120B A12B up to 262,144, Ministral 3 3B up to 262,144 tokens.
Which can read images, PDFs, audio or video?
Seed 2.0 Mini accepts text, images and video; Nemotron 3 Super 120B A12B accepts text; Ministral 3 3B accepts text and images. Seed 2.0 Mini handles the widest range of inputs.
Are any of these open source?
Nemotron 3 Super 120B A12B and Ministral 3 3B publishes its weights (Apache 2.0) and can be self-hosted; Seed 2.0 Mini is proprietary.
Which is newer?
Nemotron 3 Super 120B A12B is the newest, released Mar 11, 2026. Seed 2.0 Mini came out Feb 14, 2026; Ministral 3 3B came out Dec 2, 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.
