Apertus 8B vs Granite-4.0-H-Small vs Seed 1.6 Vision
Too close to call on our weighted score (Seed 1.6 Vision 63, Granite-4.0-H-Small 63, Apertus 8B 56). The right pick depends on what you value most.
Swiss AI
Apertus 8B
56/100- ECI—
- Price$0.10 / $0.20
- Context66K
IBMGranite-4.0-H-Small
63/100- ECI—
- Price$0.064 / $0.265
- Context131K
ByteDance SeedSeed 1.6 Vision
63/100- ECI—
- Price$0.119 / $1.19
- Context256K
Too close to call
It is close. Our weighted score puts them within a point (Seed 1.6 Vision 63/100, Granite-4.0-H-Small 63/100, Apertus 8B 56/100), so choose by what matters most for your work: Granite-4.0-H-Small on price and Seed 1.6 Vision 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 priceGranite-4.0-H-SmallGranite-4.0-H-Small $0.114 · Apertus 8B $0.125 · Seed 1.6 Vision $0.386 per 1M tokens (3:1 blend)
- Longest contextSeed 1.6 VisionSeed 1.6 Vision 256,000 · Granite-4.0-H-Small 131,072 · Apertus 8B 65,536 tokens
- Widest inputsSeed 1.6 VisionApertus 8B: Text · Granite-4.0-H-Small: Text · Seed 1.6 Vision: Text, Images
- Self-hostingApertus 8B and Granite-4.0-H-SmallPublishes downloadable weights (Apache-2.0)
| Measure | Weight | Apertus 8B | Granite-4.0-H-Small | Seed 1.6 Vision |
|---|---|---|---|---|
| Price | 50% | 93 | 95 | 69 |
| Inputs & features | 30% | 25 | 35 | 70 |
| Context window | 20% | 12 | 24 | 36 |
| Overall | 100% | 56/100 | 63/100 | 63/100 |
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 | Apertus 8BSwiss AI | Granite-4.0-H-SmallIBM | Seed 1.6 VisionByteDance Seed |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.10 | $0.064 (best) | $0.119 |
| Output | $0.20 (best) | $0.265 | $1.19 |
| Cached input | — | — | $0.024 |
| Blended (3:1) | $0.125 | $0.114 (best) | $0.386 |
| Long-context rate | Same rate | Same rate | Over 32K: $0.178 / $2.37 |
| Price source | Median of 1 providers | Official watsonx.ai API | Official Volcengine Ark API |
| Limits | |||
| Context window | 65,536 tokens | 131,072 tokens | 256,000 tokens (best) |
| Max output | 8,192 tokens | 131,072 tokens (best) | 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 | No | No | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | Yes | Yes |
| Availability | |||
| Weights | OpenApache-2.0 | Open | Proprietary |
| API model ID | — | ibm/granite-4-h-small | doubao-seed-1-6-vision-250815 |
| API providers | 1 | 1 | 2 (best) |
| Released | Sep 2, 2025 | Oct 2, 2025 | Aug 15, 2025 |
| Knowledge cutoff | Sep 2025 | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Apertus 8B$1.40
- Granite-4.0-H-Small$1.17
- Seed 1.6 Vision$3.56
Which should you choose?
Which is better: Apertus 8B, Granite-4.0-H-Small or Seed 1.6 Vision?
It is close. Our weighted score puts them within a point (Seed 1.6 Vision 63/100, Granite-4.0-H-Small 63/100, Apertus 8B 56/100), so choose by what matters most for your work: Granite-4.0-H-Small on price and Seed 1.6 Vision 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, Apertus 8B, Granite-4.0-H-Small or Seed 1.6 Vision?
Granite-4.0-H-Small is cheaper at $0.064 input / $0.265 output per million tokens (official watsonx.ai API price). Apertus 8B costs $0.10 input / $0.20 output per million tokens (median across 1 API provider); Seed 1.6 Vision costs $0.119 input / $1.19 output per million tokens (official Volcengine Ark API price). At a typical mix of three input tokens to one output token, that is $0.114 per million tokens for Granite-4.0-H-Small versus $0.125 for Apertus 8B (1.1× as much) and $0.386 for Seed 1.6 Vision (3.4× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Apertus 8B has not been scored yet, Granite-4.0-H-Small has not been scored yet and Seed 1.6 Vision has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Apertus 8B, Granite-4.0-H-Small and Seed 1.6 Vision 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?
Seed 1.6 Vision has the largest context window at 256,000 tokens, against 131,072 for Granite-4.0-H-Small and 65,536 for Apertus 8B. Maximum output per response: Apertus 8B up to 8,192, Granite-4.0-H-Small up to 131,072, Seed 1.6 Vision up to 32,000 tokens.
Which can read images, PDFs, audio or video?
Apertus 8B accepts text; Granite-4.0-H-Small accepts text; Seed 1.6 Vision accepts text and images. Seed 1.6 Vision handles the widest range of inputs.
Are any of these open source?
Apertus 8B and Granite-4.0-H-Small publishes its weights (Apache-2.0) and can be self-hosted; Seed 1.6 Vision is proprietary.
Which is newer?
Granite-4.0-H-Small is the newest, released Oct 2, 2025. Apertus 8B came out Sep 2, 2025; Seed 1.6 Vision came out Aug 15, 2025. Knowledge cutoff: Apertus 8B Sep 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.
