MolmoAct2 Pretrain vs Hy4 preview
Benchmark Performance
Available Benchmarks
Technical Differences
Side-by-Side Facts
| Field | MolmoAct2-Pretrain | Hy4 preview |
|---|---|---|
| Developer | Ai2 | Tencent |
| Family | Molmoact2 Pretrain | Hy4 |
| Model | MolmoAct2-Pretrain | Hy4 preview |
| Version | MolmoAct2-Pretrain | Hy4 preview |
| Lifecycle | active | preview |
| Released | Unknown | 2026-08-28 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Model-specific input | Text |
| Output modalities | Model-specific input | Text |
| Context window | 16K | 1,000K |
| Total parameters | 4.9B | 770B |
| Active parameters | Unknown | 49B |
| License | Unknown | apache-2.0 |
| Open weights | Yes | Yes |
| API available | Unknown | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Unknown | Openrouter (Standard) |
| Capabilities | robot_action_generation | chat, generation, reasoning, tools |
12 comparable fields · 10 material differences · Pair passes the primary-source comparison gate
MolmoAct2 Pretrain Capabilities
Hy4 preview Capabilities
Internal Comparison Graph
Related Comparisons
| A | Pair | B | Context |
|---|---|---|---|
| vs |  GPT-5.6 SolOpenAI | cross-developer peersimage, text | |
| vs | Claude Fable 5.1Anthropic | cross-developer peersimage, text | |
| vs |  Gemini 3.1 ProGoogle DeepMind | cross-developer peerstext | |
| vs |  DeepSeek-V4-ProDeepSeek | cross-developer peerstext | |
| vs | Grok 4.6xAI | cross-developer peersimage, text | |
| vs |  Qwen3.8-MaxQwen | cross-developer peersimage, text | |
| vs |  Kimi-K3Moonshot AI | cross-developer peersimage, text | |
 MiniMax-M3MiniMax | vs | cross-developer peersimage, text | |
| vs |  GLM-5.3Z.ai | cross-developer peerstext | |
| vs | Seed 2.1 ProByteDance Seed | cross-developer peersimage, text | |
| vs |  Mistral Large 3Mistral AI | cross-developer peersimage, text | |
| vs | cross-developer peersimage, text |
Primary Evidence
Sources and Freshness
Questions
MolmoAct2 Pretrain vs Hy4 preview FAQs
Is MolmoAct2 Pretrain or Hy4 preview better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both MolmoAct2 Pretrain and Hy4 preview, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, MolmoAct2 Pretrain or Hy4 preview?+
Only Hy4 preview has a directly sourced input price: $0.834 per million tokens. Only Hy4 preview has a directly sourced output price: $2.501 per million tokens.
Which has a larger context window, MolmoAct2 Pretrain or Hy4 preview?+
Hy4 preview has the larger sourced context window. MolmoAct2 Pretrain supports 16K and Hy4 preview supports 1,000K.
Which performs better in benchmarks, MolmoAct2 Pretrain or Hy4 preview?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can MolmoAct2 Pretrain or Hy4 preview be self-hosted?+
Both models have the same recorded self-hosting status: supported. MolmoAct2 Pretrain is open weight; Hy4 preview is open weight.
Can MolmoAct2 Pretrain and Hy4 preview understand images?+
MolmoAct2 Pretrain is not documented with image input; Hy4 preview is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, MolmoAct2 Pretrain or Hy4 preview?+
Neither has a larger sourced maximum output. MolmoAct2 Pretrain is — and Hy4 preview is —.
Do MolmoAct2 Pretrain and Hy4 preview support reasoning and tool use?+
MolmoAct2 Pretrain: none of these features are definitively sourced. Hy4 preview: reasoning and tool calling. Feature support does not establish relative quality.
Which is available from more inference providers, MolmoAct2 Pretrain or Hy4 preview?+
MolmoAct2 Pretrain has 0 sourced provider routes; Hy4 preview has 1, so Hy4 preview has broader tracked availability.
Which offers better value, MolmoAct2 Pretrain or Hy4 preview?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.