Ember-1 vs Qwen 3.8 Max Prime vs SenseNova-Loop-9B
No independent capability benchmark covers every model here yet. The score uses context window.
Price in rating: Included · Exclude price
Fireworks AI
Ember-1
65/100- ECI—
- Price$3.00 / $15.00
- Context1.05M
Alibaba (Qwen)
Qwen 3.8 Max Prime
64/100- ECI—
- Price$4.00 / $12.00
- Context1M
SenseTime
SenseNova-Loop-9B
54/100- ECI—
- Price—
- Context524K
Capability evidence is incomplete
There is not enough shared capability evidence to name an overall winner. The scores (Ember-1 65/100, Qwen 3.8 Max Prime 64/100, SenseNova-Loop-9B 54/100) compare the available specifications and price. The score weighs context window 100%. 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 priceEmber-1 and Qwen 3.8 Max PrimeEmber-1 $6.00 · Qwen 3.8 Max Prime $6.00 per 1M tokens (3:1 blend) · SenseNova-Loop-9B unpriced
- Longest contextEmber-1Ember-1 1,048,576 · Qwen 3.8 Max Prime 1,000,000 · SenseNova-Loop-9B 524,288 tokens
- Widest inputsQwen 3.8 Max PrimeEmber-1: Text, Images · Qwen 3.8 Max Prime: Text, Images, Video · SenseNova-Loop-9B: Text, Images
- Self-hostingSenseNova-Loop-9BPublishes downloadable weights
| Measure | Weight | Ember-1 | Qwen 3.8 Max Prime | SenseNova-Loop-9B |
|---|---|---|---|---|
| Context window | 100% | 65 | 64 | 54 |
| Specs & price only | 100% | 65/100 | 64/100 | 54/100 |
Left out because at least one model lacks the data: capability, price and inputs & features. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Task | — | — | Vision-language reasoning |
| Published capabilities | Specialized model from Fireworks built on Kimi K3, producing shorter reasoning traces with approximately 40% fewer tokens while maintaining comparable quality | High-throughput edition of Qwen3.8 Max for coding, professional work, multimodal understanding, and long-running agent workflows | Reuses middle layers to perform 56 layer executions with 32 physical decoder layers. |
| Model size | — | — | 9B parameters |
| Access | — | — | Public weights including quantized variants for local inference |
| Sources | — | — | Model weights |
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $3.00 (best) | $4.00 | — |
| Output | $15.00 | $12.00 (best) | — |
| Cached input | — | — | — |
| Blended (3:1) | $6.00 | $6.00 | — |
| Long-context rate | Same rate | Same rate | — |
| Price source | Median of 8 providers | Median of 6 providers | — |
| Limits | |||
| Context window | 1,048,576 tokens (best) | 1,000,000 tokens | 524,288 tokens |
| Max output | 131,072 tokens | 131,072 tokens | — |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | Yes | Yes | Unconfirmed |
| Tool calling | Yes | Yes | — |
| Structured output | No | No | — |
| Availability | |||
| Weights | Proprietary | Proprietary | Open |
| API model ID | — | — | — |
| API providers | 8 (best) | 6 | — |
| Released | Sep 22, 2026 | Sep 23, 2026 | Nov 17, 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.
Ember-1$60.00
Qwen 3.8 Max Prime$64.00
SenseNova-Loop-9B—
Which should you choose?
Which is better: Ember-1, Qwen 3.8 Max Prime or SenseNova-Loop-9B?
There is not enough shared capability evidence to name an overall winner. The scores (Ember-1 65/100, Qwen 3.8 Max Prime 64/100, SenseNova-Loop-9B 54/100) compare the available specifications and price. The score weighs context window 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Ember-1, Qwen 3.8 Max Prime or SenseNova-Loop-9B?
Ember-1 is cheaper at $3.00 input / $15.00 output per million tokens (median across 8 API providers). Qwen 3.8 Max Prime costs $4.00 input / $12.00 output per million tokens (median across 6 API providers). At a typical mix of three input tokens to one output token, that is $6.00 per million tokens for Ember-1 versus $6.00 for Qwen 3.8 Max Prime (1× as much). SenseNova-Loop-9B has no published per-token price.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Ember-1 has not been scored yet, Qwen 3.8 Max Prime has not been scored yet and SenseNova-Loop-9B has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Ember-1, Qwen 3.8 Max Prime and SenseNova-Loop-9B yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that SenseNova-Loop-9B has no confirmed tool-calling support in this comparison.
Which has the bigger context window?
Ember-1 has the largest context window at 1,048,576 tokens, against 1,000,000 for Qwen 3.8 Max Prime and 524,288 for SenseNova-Loop-9B. Maximum output per response: Ember-1 up to 131,072, Qwen 3.8 Max Prime up to 131,072 tokens.
Which can read images, PDFs, audio or video?
Ember-1 accepts text and images; Qwen 3.8 Max Prime accepts text, images and video; SenseNova-Loop-9B accepts text and images. Qwen 3.8 Max Prime handles the widest range of inputs.
Are any of these open source?
SenseNova-Loop-9B publishes its weights and can be self-hosted; Ember-1 and Qwen 3.8 Max Prime is proprietary.
Which is newer?
SenseNova-Loop-9B is the newest, released Nov 17, 2026. Qwen 3.8 Max Prime came out Sep 23, 2026; Ember-1 came out Sep 22, 2026.
How do you decide the winner?
This rating uses context window (100%, 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 19, 2026.