# Atria Dawn Preview

Shanghai Artificial Intelligence Laboratory · Open Weight · bench-align-v5 · Self-hosted

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/models/atria-dawn-preview  
JSON: https://modelscale.dev/api/model/atria-dawn-preview

## Facts

| Field | Value | Evidence |
| --- | --- | --- |
| Canonical id | `atria-dawn-preview` | — |
| Overall score | Unavailable | Unavailable · source benchlm:models |
| Verified score (BenchLM): a separate BenchLM figure, not the overall score | 68 | Observed 2026-11-21 · source benchlm:models |
| Context window | 256K tokens | Observed 2026-11-21 · source benchlm:models |
| Release date | 2026-09-14 | Observed 2026-11-21 · source benchlm:models |
| Access type | Open Weight | — |
| Blended $/1M (75% input / 25% output) | Unavailable | Derived from the input and output rates below |

**Only a verified score is published.** BenchLM publishes no overall score for this model, so the overall score stays unavailable. The verified score beside it is a separate figure in BenchLM's data. BenchLM does not document how the two are calculated differently, and they can differ widely, so one is never used in place of the other.

## Capability evidence

Seven axes from the ranking source. An axis the source did not score is unavailable, not zero. Beside each score is BenchLM's separate verified score for that axis; the two are published side by side and are not interchangeable.

| Axis | Score | Evidence | Verified (BenchLM) | Verified evidence |
| --- | --- | --- | --- | --- |
| Agentic | 82.8 | Observed 2026-11-21 · source benchlm:models | 84.5 | Observed 2026-11-21 · source benchlm:models |
| Coding | 48.3 | Observed 2026-11-21 · source benchlm:models | 48.3 | Observed 2026-11-21 · source benchlm:models |
| Knowledge | Unavailable | Unavailable · source benchlm:models | Unavailable | Unavailable · source benchlm:models |
| Reasoning | Unavailable | Unavailable · source benchlm:models | Unavailable | Unavailable · source benchlm:models |
| Multimodal & Grounded | Unavailable | Unavailable · source benchlm:models | Unavailable | Unavailable · source benchlm:models |
| Instruction Following | Unavailable | Unavailable · source benchlm:models | Unavailable | Unavailable · source benchlm:models |
| Math | Unavailable | Unavailable · source benchlm:models | Unavailable | Unavailable · source benchlm:models |

## Runtime service evidence

Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim. Regional or per-endpoint measurements appear only when the API supplies them; none are modelled.

| Measurement | Value | Observed | Last good | Evidence |
| --- | --- | --- | --- | --- |
| Time to first token | Unavailable | Unavailable | Unavailable | Unavailable · source benchlm:speed |
| Throughput | Unavailable | Unavailable | Unavailable | Unavailable · source benchlm:speed |

## Endpoint and price matrix

Every published price component, including cache reads and writes.

| Component | USD | Evidence |
| --- | --- | --- |
| Input / 1M tokens | Unavailable | Unavailable · source benchlm:pricing |
| Output / 1M tokens | Unavailable | Unavailable · source benchlm:pricing |
| Cache read / 1M tokens | Unavailable | Unavailable · source benchlm:pricing |
| Cache write / 1M tokens | Unavailable | Unavailable · source openrouter:pricing |
| Blended / 1M (75% input / 25% output) | Unavailable | Derived — from the input and output rates above; it has no source record of its own |
| Cost per successful task (LiveBench) | Unavailable | Unavailable · source livebench:table |

**Self-hosted listing.** Shanghai Artificial Intelligence Laboratory publishes the full-precision and FP8 Atria Dawn Preview checkpoints under MIT for self-hosting. The pricing catalog represents the open-weight row as self-host/free-per-token before infrastructure costs. Atria also offers the exact `Atria-Dawn-Preview` model through its hosted API, but its public site and API documentation do not publish a first-party token rate. No hosted token rate was published for this model, so its per-token price is unavailable rather than zero.

## Workload-aware monthly cost example

10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. Derived here from the published rates above by this site's own calculator — not a figure any source published.

| Field | Value |
| --- | --- |
| Modelled monthly cost | Unavailable |
| Modelled tokens | Unavailable |
| Reason | The applicable input rate is unavailable. |

## Benchmark record

13 matched benchmark rows with their published value, unit, and provenance.

### Coding

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| Terminal-Bench 2.1 | 78.3 | Terminal-based software-agent tasks | Professional software engineering | [Terminal-Bench](https://www.tbench.ai/) |
| SWE-bench Pro | 59.6 | 1,865 repository problems | Long-horizon professional engineering | [SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?](https://arxiv.org/abs/2509.16941) |

### Agentic

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| Terminal-Bench 2.1 | 78.3 | Terminal-based software-agent tasks | Professional software engineering | [Terminal-Bench](https://www.tbench.ai/) |
| BrowseComp | 92.5 | Research questions requiring browsing | Hard web research | [BrowseComp](https://openai.com/index/browsecomp/) |
| GDPval-AA | 1583 | Agentic real-world work tasks | Professional agentic workflows | [DeepSeek-V4 Technical Report](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek_V4.pdf) |
| CyberGym | 86.5 | 1,507 vulnerability analysis instances | Real-world cybersecurity | [CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale](https://www.cybergym.io/) |
| JobBench | 50.3 | 130 tasks across 35 occupations | Professional multi-source workflows | [JobBench: Aligning Agent Work With Human Will](https://arxiv.org/abs/2605.26329) |
| AutomationBench | 53.8 | 600 public automation tasks | Long-horizon automation | [Kimi K3: Open Frontier Intelligence](https://www.kimi.com/blog/kimi-k3) |
| DeepSearchQA | 96.0 | Agentic browsing and list-answer questions | Agentic web research | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |
| BFCL v4 (Berkeley Function Calling Leaderboard v4) | 77.0 | Function-calling tasks | Advanced tool use | [The Berkeley Function Calling Leaderboard (BFCL): From Tool Use to Agentic Evaluation of Large Language Models](https://gorilla.cs.berkeley.edu/leaderboard.html) |
| MLE-Bench Lite | 86.2 | Low-resource ML competitions | Agentic machine learning | [MiniMax M2.7: Early Echoes of Self-Evolution](https://www.minimax.io/news/minimax-m27-en) |
| τ³-bench results (τ³-Bench Tool-Agent-User Evaluation) | 41.2 | Corrected customer-service tasks plus knowledge and voice evaluation modes | Long-horizon, multimodal, and knowledge-aware tool use | [Official τ³-bench repository and release notes](https://github.com/sierra-research/tau2-bench) |
| WideResearch | 81.9 | Open-ended research tasks | Broad research-agent workflows | [Qwen3.6 launch benchmarks](https://qwen.ai/blog?id=qwen3.6) |

## Lifecycle and limitations log

Lifecycle events the source associates with this model.

No lifecycle event references this model. That is not evidence the model has no lifecycle plan — only that this source published none.

## What this profile does not claim

Values are reproduced exactly as their sources published them, in the units those sources declared; none are converted, interpolated, or averaged across providers. Any field marked unavailable was attempted and not returned. Last attempted fetch for this model's score: 2026-11-21 01:17 UTC.
