Step 3.7 Flash vs DeepSeek V4 Flash Vision Exp vs Nemotron 3 Nano Omni 30B A3B Reasoning
Too close to call on our weighted score (DeepSeek V4 Flash Vision Exp 70, Nemotron 3 Nano Omni 30B A3B Reasoning 69, Step 3.7 Flash 62). The right pick depends on what you value most.
StepFunStep 3.7 Flash
62/100- ECI—
- Price$0.185 / $1.11
- Context256K
DeepSeekDeepSeek V4 Flash Vision Exp
70/100- ECI—
- Price$0.216 / $0.647
- Context1M
NVIDIANemotron 3 Nano Omni 30B A3B Reasoning
69/100- ECI—
- Price$0.25 / $0.85
- Context256K
Too close to call
It is close. Our weighted score puts them within a point (DeepSeek V4 Flash Vision Exp 70/100, Nemotron 3 Nano Omni 30B A3B Reasoning 69/100, Step 3.7 Flash 62/100), so choose by what matters most for your work: DeepSeek V4 Flash Vision Exp 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 priceDeepSeek V4 Flash Vision ExpDeepSeek V4 Flash Vision Exp $0.323 · Nemotron 3 Nano Omni 30B A3B Reasoning $0.40 · Step 3.7 Flash $0.416 per 1M tokens (3:1 blend)
- Longest contextDeepSeek V4 Flash Vision ExpDeepSeek V4 Flash Vision Exp 1,000,000 · Step 3.7 Flash 256,000 · Nemotron 3 Nano Omni 30B A3B Reasoning 256,000 tokens
- Widest inputsNemotron 3 Nano Omni 30B A3B ReasoningStep 3.7 Flash: Text, Images, Video · DeepSeek V4 Flash Vision Exp: Text, Images · Nemotron 3 Nano Omni 30B A3B Reasoning: Text, Images, Audio, Video
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Step 3.7 Flash | DeepSeek V4 Flash Vision Exp | Nemotron 3 Nano Omni 30B A3B Reasoning |
|---|---|---|---|---|
| Price | 50% | 68 | 73 | 69 |
| Inputs & features | 30% | 70 | 70 | 90 |
| Context window | 20% | 36 | 60 | 36 |
| Overall | 100% | 62/100 | 70/100 | 69/100 |
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 | Step 3.7 FlashStepFun | DeepSeek V4 Flash Vision ExpDeepSeek | Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.185 (best) | $0.216 | $0.25 |
| Output | $1.11 | $0.647 (best) | $0.85 |
| Cached input | $0.037 | — | — |
| Blended (3:1) | $0.416 | $0.323 (best) | $0.40 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official StepFun (Global) API | Median of 15 providers | Median of 4 providers |
| Limits | |||
| Context window | 256,000 tokens | 1,000,000 tokens (best) | 256,000 tokens |
| Max output | 256,000 tokens | 384,000 tokens (best) | 65,536 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | No | No | No |
| Audio | No | No | Yes |
| Video | Yes | No | Yes |
| Reasoning | Yeslow · medium · high | Yes | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | Yes | Yes |
| Availability | |||
| Weights | Open | OpenMIT | Open |
| API model ID | step-3.7-flash | — | nvidia/nemotron-3-nano-omni-30b-a3b-reasoning |
| API providers | 18 (best) | 15 | 8 |
| Released | May 29, 2026 | Aug 21, 2026 | Apr 28, 2026 |
| Knowledge cutoff | Mar 1, 2026 | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Step 3.7 Flash$4.07
- DeepSeek V4 Flash Vision Exp$3.45
- Nemotron 3 Nano Omni 30B A3B Reasoning$4.20
Which should you choose?
Which is better: Step 3.7 Flash, DeepSeek V4 Flash Vision Exp or Nemotron 3 Nano Omni 30B A3B Reasoning?
It is close. Our weighted score puts them within a point (DeepSeek V4 Flash Vision Exp 70/100, Nemotron 3 Nano Omni 30B A3B Reasoning 69/100, Step 3.7 Flash 62/100), so choose by what matters most for your work: DeepSeek V4 Flash Vision Exp 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, Step 3.7 Flash, DeepSeek V4 Flash Vision Exp or Nemotron 3 Nano Omni 30B A3B Reasoning?
DeepSeek V4 Flash Vision Exp is cheaper at $0.216 input / $0.647 output per million tokens (median across 15 API providers). Nemotron 3 Nano Omni 30B A3B Reasoning costs $0.25 input / $0.85 output per million tokens (median across 4 API providers; free on Nvidia); Step 3.7 Flash costs $0.185 input / $1.11 output per million tokens (official StepFun (Global) API price). At a typical mix of three input tokens to one output token, that is $0.323 per million tokens for DeepSeek V4 Flash Vision Exp versus $0.40 for Nemotron 3 Nano Omni 30B A3B Reasoning (1.2× as much) and $0.416 for Step 3.7 Flash (1.3× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Step 3.7 Flash has not been scored yet, DeepSeek V4 Flash Vision Exp has not been scored yet and Nemotron 3 Nano Omni 30B A3B Reasoning has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Step 3.7 Flash, DeepSeek V4 Flash Vision Exp and Nemotron 3 Nano Omni 30B A3B Reasoning 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 256,000 for Step 3.7 Flash and 256,000 for Nemotron 3 Nano Omni 30B A3B Reasoning. Maximum output per response: Step 3.7 Flash up to 256,000, DeepSeek V4 Flash Vision Exp up to 384,000, Nemotron 3 Nano Omni 30B A3B Reasoning up to 65,536 tokens.
Which can read images, PDFs, audio or video?
Step 3.7 Flash accepts text, images and video; DeepSeek V4 Flash Vision Exp accepts text and images; Nemotron 3 Nano Omni 30B A3B Reasoning accepts text, images, audio and video. Nemotron 3 Nano Omni 30B A3B Reasoning handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights (MIT), so you can self-host them.
Which is newer?
DeepSeek V4 Flash Vision Exp is the newest, released Aug 21, 2026. Step 3.7 Flash came out May 29, 2026; Nemotron 3 Nano Omni 30B A3B Reasoning came out Apr 28, 2026. Knowledge cutoff: Step 3.7 Flash Mar 1, 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.
