Inkling vs Qwen3.8 27B vs Seed 2.1 Pro
No independent capability benchmark covers every model here yet. The score uses price, inputs & features and context window.
Price in rating: Included · Exclude price
Thinking Machines
Inkling
49/100- ECI148.4
- Price$1.87 / $4.68
- Context1.05M
Alibaba (Qwen)
Qwen3.8 27B
54/100- ECI152.7
- Price$0.40 / $2.42
- Context262K
ByteDance Seed
Seed 2.1 Pro
55/100- ECI—
- Price$0.891 / $4.45
- Context1.02M
Capability evidence is incomplete
There is not enough shared capability evidence to name an overall winner. The scores (Seed 2.1 Pro 55/100, Qwen3.8 27B 54/100, Inkling 49/100) compare the available specifications and price. 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 priceQwen3.8 27BQwen3.8 27B $0.905 · Seed 2.1 Pro $1.78 · Inkling $2.57 per 1M tokens (3:1 blend)
- Longest contextInklingInkling 1,048,576 · Seed 2.1 Pro 1,024,000 · Qwen3.8 27B 262,144 tokens
- Widest inputsSame inputsInkling: Text, Images, Audio · Qwen3.8 27B: Text, Images, Video · Seed 2.1 Pro: Text, Images, Video
- Self-hostingInkling and Qwen3.8 27BPublishes downloadable weights (Apache-2.0)
| Measure | Weight | Inkling | Qwen3.8 27B | Seed 2.1 Pro |
|---|---|---|---|---|
| Price | 50% | 29 | 50 | 36 |
| Inputs & features | 30% | 70 | 70 | 80 |
| Context window | 20% | 65 | 43 | 65 |
| Specs & price only | 100% | 49/100 | 54/100 | 55/100 |
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 | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 148.4 | 152.7 (best) | — |
| ECI rank | #74 of 158 | #57 of 158 (best) | — |
| GPQA DiamondGraduate-level science questions | 88.3% | — | 89.3% (best) |
| FrontierMath Tiers 1–3Research-level mathematics | 33.3% | — | — |
| OTIS Mock AIME 2024–2025Competition mathematics | 88.9% | — | 95.6% (best) |
| SimpleQA VerifiedShort factual questions | 40.3% | — | 44.7% (best) |
| Price per million tokens | |||
| Input | $1.87 | $0.40 (best) | $0.891 |
| Output | $4.68 | $2.42 (best) | $4.45 |
| Cached input | $0.374 | — | $0.178 (best) |
| Blended (3:1) | $2.57 | $0.905 (best) | $1.78 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Thinking Machines API | Median of 67 providers | Official Volcengine Ark API |
| Limits | |||
| Context window | 1,048,576 tokens (best) | 262,144 tokens | 1,024,000 tokens |
| Max output | 1,048,576 tokens (best) | 32,768 tokens | 256,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | No | No | No |
| Audio | Yes | No | No |
| Video | No | Yes | Yes |
| Reasoning | Yeslow · medium · high · xhigh · max | Yes | Yesminimal · low · medium · high · xhigh · max |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | Yes |
| Availability | |||
| Weights | OpenApache-2.0 | Open | Proprietary |
| API model ID | thinkingmachines/Inkling | — | doubao-seed-2-1-pro-260628 |
| API providers | 18 | 72 (best) | 3 |
| Released | Jul 15, 2026 | Aug 14, 2026 | Jun 23, 2026 |
| 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.
Inkling$28.06
Qwen3.8 27B$8.84
Seed 2.1 Pro$17.81
Which should you choose?
Which is better: Inkling, Qwen3.8 27B or Seed 2.1 Pro?
There is not enough shared capability evidence to name an overall winner. The scores (Seed 2.1 Pro 55/100, Qwen3.8 27B 54/100, Inkling 49/100) compare the available specifications and price. 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, Inkling, Qwen3.8 27B or Seed 2.1 Pro?
Qwen3.8 27B is cheaper at $0.40 input / $2.42 output per million tokens (median across 67 API providers). Seed 2.1 Pro costs $0.891 input / $4.45 output per million tokens (official Volcengine Ark API price); Inkling costs $1.87 input / $4.68 output per million tokens (official Thinking Machines API price). At a typical mix of three input tokens to one output token, that is $0.905 per million tokens for Qwen3.8 27B versus $1.78 for Seed 2.1 Pro (2× as much) and $2.57 for Inkling (2.8× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Inkling has an ECI of 148.4, Qwen3.8 27B has an ECI of 152.7 and Seed 2.1 Pro has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Inkling, Qwen3.8 27B and Seed 2.1 Pro 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?
Inkling has the largest context window at 1,048,576 tokens, against 1,024,000 for Seed 2.1 Pro and 262,144 for Qwen3.8 27B. Maximum output per response: Inkling up to 1,048,576, Qwen3.8 27B up to 32,768, Seed 2.1 Pro up to 256,000 tokens.
Which can read images, PDFs, audio or video?
Inkling accepts text, images and audio; Qwen3.8 27B accepts text, images and video; Seed 2.1 Pro accepts text, images and video. They handle the same number of input types.
Are any of these open source?
Inkling and Qwen3.8 27B publishes its weights (Apache-2.0) and can be self-hosted; Seed 2.1 Pro is proprietary.
Which is newer?
Qwen3.8 27B is the newest, released Aug 14, 2026. Inkling came out Jul 15, 2026; Seed 2.1 Pro came out Jun 23, 2026.
How do you decide the winner?
This rating uses price (50%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (30%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (20%, 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 Nov 18, 2026.