Nemotron 3.5 Lightning 30B A3B vs Sarvam 105B vs Seed 1.6 Flash
Seed 1.6 Flash comes out ahead, 78 to 71 and 65 on our weighted score, and it is the cheaper option too.
NVIDIANemotron 3.5 Lightning 30B A3B
71/100- ECI—
- Price$0.05 / $0.20
- Context262K
Sarvam AI
Sarvam 105B
65/100- ECI—
- Price$0.047 / $0.186
- Context131K
- Our pick
ByteDance SeedSeed 1.6 Flash
78/100- ECI—
- Price$0.022 / $0.223
- Context256K
Seed 1.6 Flash is our pick
Seed 1.6 Flash is the better all-round choice, scoring 78/100 against Nemotron 3.5 Lightning 30B A3B (71) and Sarvam 105B (65). 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 1.6 FlashSeed 1.6 Flash $0.072 · Sarvam 105B $0.082 · Nemotron 3.5 Lightning 30B A3B $0.087 per 1M tokens (3:1 blend)
- Longest contextNemotron 3.5 Lightning 30B A3BNemotron 3.5 Lightning 30B A3B 262,144 · Seed 1.6 Flash 256,000 · Sarvam 105B 131,072 tokens
- Widest inputsSeed 1.6 FlashNemotron 3.5 Lightning 30B A3B: Text · Sarvam 105B: Text · Seed 1.6 Flash: Text, Images
- Self-hostingNemotron 3.5 Lightning 30B A3B and Sarvam 105BPublishes downloadable weights
| Measure | Weight | Nemotron 3.5 Lightning 30B A3B | Sarvam 105B | Seed 1.6 Flash |
|---|---|---|---|---|
| Price | 50% | 100 | 100 | 100 |
| Inputs & features | 30% | 45 | 35 | 70 |
| Context window | 20% | 37 | 24 | 36 |
| Overall | 100% | 71/100 | 65/100 | 78/100 |
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 | Nemotron 3.5 Lightning 30B A3BNVIDIA | Sarvam 105BSarvam AI | Seed 1.6 FlashByteDance Seed |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.05 | $0.047 | $0.022 (best) |
| Output | $0.20 | $0.186 (best) | $0.223 |
| Cached input | — | — | $0.0044 |
| Blended (3:1) | $0.087 | $0.082 | $0.072 (best) |
| Long-context rate | Same rate | Same rate | Over 32K: $0.045 / $0.445 |
| Price source | Median of 9 providers | Median of 2 providers | Official Volcengine Ark API |
| Limits | |||
| Context window | 262,144 tokens (best) | 131,072 tokens | 256,000 tokens |
| Max output | 262,144 tokens (best) | 131,072 tokens | 32,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yes · low · medium · high | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | No | Yes |
| Availability | |||
| Weights | Open | Open | Proprietary |
| API model ID | nvidia/nemotron-3.5-lightning-30b-a3b | sarvam-105b | doubao-seed-1-6-flash-250828 |
| API providers | 12 (best) | 3 | 2 |
| Released | Aug 11, 2026 | Sep 1, 2025 | Aug 28, 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.
- Nemotron 3.5 Lightning 30B A3B$0.90
- Sarvam 105B$0.842
- Seed 1.6 Flash$0.668
Which should you choose?
Which is better: Nemotron 3.5 Lightning 30B A3B, Sarvam 105B or Seed 1.6 Flash?
Seed 1.6 Flash is the better all-round choice, scoring 78/100 against Nemotron 3.5 Lightning 30B A3B (71) and Sarvam 105B (65). 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, Nemotron 3.5 Lightning 30B A3B, Sarvam 105B or Seed 1.6 Flash?
Seed 1.6 Flash is cheaper at $0.022 input / $0.223 output per million tokens (official Volcengine Ark API price). Sarvam 105B costs $0.047 input / $0.186 output per million tokens (median across 2 API providers); Nemotron 3.5 Lightning 30B A3B costs $0.05 input / $0.20 output per million tokens (median across 9 API providers; free on Nvidia). At a typical mix of three input tokens to one output token, that is $0.072 per million tokens for Seed 1.6 Flash versus $0.082 for Sarvam 105B (1.1× as much) and $0.087 for Nemotron 3.5 Lightning 30B A3B (1.2× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Nemotron 3.5 Lightning 30B A3B has not been scored yet, Sarvam 105B has not been scored yet and Seed 1.6 Flash has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Nemotron 3.5 Lightning 30B A3B, Sarvam 105B and Seed 1.6 Flash 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 3.5 Lightning 30B A3B has the largest context window at 262,144 tokens, against 256,000 for Seed 1.6 Flash and 131,072 for Sarvam 105B. Maximum output per response: Nemotron 3.5 Lightning 30B A3B up to 262,144, Sarvam 105B up to 131,072, Seed 1.6 Flash up to 32,000 tokens.
Which can read images, PDFs, audio or video?
Nemotron 3.5 Lightning 30B A3B accepts text; Sarvam 105B accepts text; Seed 1.6 Flash accepts text and images. Seed 1.6 Flash handles the widest range of inputs.
Are any of these open source?
Nemotron 3.5 Lightning 30B A3B and Sarvam 105B publishes its weights and can be self-hosted; Seed 1.6 Flash is proprietary.
Which is newer?
Nemotron 3.5 Lightning 30B A3B is the newest, released Aug 11, 2026. Sarvam 105B came out Sep 1, 2025; Seed 1.6 Flash came out Aug 28, 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.
