Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
Scroll to compare models →
| 2 models | QwenQwen3.8-27BRemove × | metaMeta: Muse Spark OpenRemove × |
|---|---|---|
| What’s different | Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking… | OpenRouter lists a 1,048,576-token context and up to 943,718 output tokens for this open-weight model aimed at agentic coding. |
| Date | Aug 14, 2026Released | Oct 15, 2026Released |
API prices USD · provider configuration shown below | ||
| Listed rate | $0.039 in / $1.25 out/ 1M tokensChutes · via OpenRouter · all routes ↗chutes/fp8 · quantization=fp8 | $1.25 in / $4.25 out/ 1M tokensbaseten · via Hugging Face · all routes ↗ |
| Provider endpoints | 28 endpoints → | 8 endpoints → |
Benchmarks Each result retains its evaluation configuration | ||
| AA Intelligence ↗4.3.2 | 33.68 ↗Effort: xhigh · XhighIndependent result | 38.87 ↗Effort: xhigh · XhighIndependent result |
| AA-Briefcase ↗1.1 | 1,394.92 ↗Effort: xhigh · XhighIndependent result | 1,279.37 ↗Effort: xhigh · XhighIndependent result |
| GDPval-AA ↗2.1 | 1,416.92 ↗Effort: xhigh · XhighIndependent result | 1,401.12 ↗Effort: xhigh · XhighIndependent result |
| AutomationBench-AA ↗AA Index 4.3.2 collection | 48.24% ↗Effort: xhigh · XhighIndependent result | 49.71% ↗Effort: xhigh · XhighIndependent result |
| Terminal-Bench 4.0 · AA ↗ | 5.56% ↗Effort: xhigh · XhighHarness unreportedIndependent result | 8.08% ↗Effort: xhigh · XhighHarness unreportedIndependent result |
| SciCode ↗AA Index 4.3.2 collection | 46.64% ↗AA · Under reviewEffort: xhigh · XhighIndependent result | 58.45% ↗AA · Under reviewEffort: xhigh · XhighIndependent result |
| Humanity’s Last Exam ↗AA Index 4.3.2 collection | 33.92% ↗Effort: xhigh · XhighIndependent result | 43.88% ↗Effort: xhigh · XhighIndependent result |
| GDP.pdf ↗AA Index 4.3.2 collection | 16.4% ↗Effort: xhigh · XhighIndependent result | 18.6% ↗Effort: xhigh · XhighIndependent result |
| CritPt ↗AA Index 4.3.2 collection | 5.43% ↗AA · Under reviewEffort: xhigh · XhighIndependent result | 14.86% ↗AA · Under reviewEffort: xhigh · XhighIndependent result |
| AA-Omniscience ↗AA Index 4.3.2 collection | -9.98 ↗Effort: xhigh · XhighIndependent result | 24.65 ↗Effort: xhigh · XhighIndependent result |
| AA-LCR ↗1.1 | 82% ↗Effort: xhigh · XhighIndependent result | 78.33% ↗Effort: xhigh · XhighIndependent result |
| LiveBench ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 62.66% ↗Effort: max · single run · equal mean of 7 suitesIndependent result | 57.57% ↗Effort: max · single run · equal mean of 7 suitesIndependent result |
| LiveBench Design ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 75.9% ↗Effort: max · single runIndependent result | 60.3% ↗Effort: max · single runIndependent result |
| LiveBench Research ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 59.4% ↗Effort: max · single runIndependent result | 80.3% ↗Effort: max · single runIndependent result |
| LiveBench Browser Use ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 32% ↗Effort: max · single runIndependent result | 36% ↗Effort: max · single runIndependent result |
| LiveBench Coding ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 86.8% ↗Effort: max · single runIndependent result | 73.6% ↗Effort: max · single runIndependent result |
| LiveBench Agentic ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 27.9% ↗Effort: max · single runIndependent result | 18.9% ↗Effort: max · single runIndependent result |
| LiveBench Data Analysis ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 76.6% ↗Effort: max · single runIndependent result | 60.2% ↗Effort: max · single runIndependent result |
| LiveBench Reasoning ↗2.0 · 2026-10-14 (dataset 2026-09-02) | 80% ↗Effort: max · single runIndependent result | 73.7% ↗Effort: max · single runIndependent result |
Specifications | ||
| Context window | 1M | 1M |
| Max output | 235.9K | 943.7K |
| Parameters | 27B | — |
| Tool calling | Supported | Supported |
| Structured output | Supported | Supported |
| Image input | Supported | Supported |
| Audio input | Not reported | Supported |
| Video input | Supported | Supported |
| Reasoning | Supported | Supported |
| Prompt caching | Supported | Supported |
| Open weights | Supported | Supported |