The Orchestrator › Models › Particle AI
Particle AI: Particle-1
Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data...
Specifications
| Provider | Particle AI |
| Model ID | particle-ai/particle-1 |
| Context window | 1.0M tokens (1,048,576) |
| Modalities | text + image → text |
| Tokenizer | Other |
| Router sources | OpenRouter, Vercel AI Gateway, Cloudflare |
| Released | 2026-10-28 |
| Tracked since | 2026-11-14 |
| Popularity | Not in the ranked set |
| Reasoning | On by default |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | particle-ai/particle-1 | $0.05 | $0.5 | 1.0M |
| free | OpenRouter | particle-ai/particle-1:free | Free | Free | 1.0M |
| standard | Vercel AI Gateway | particle-ai/particle-1 | $0.125 | $0.5 | 1.0M |
| standard | Cloudflare | particle-ai/particle-1 | $0.25 | $1.00 | — |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.25 | $1.00 | — |
| 2026-11-14 | $0.125 | $0.5 | 1.0M |
| 2026-11-14 | $0.05 | $0.5 | 1.0M |
Capabilities
Vision (image input)Agentic tool useStructured outputsReasoning mode1M+ context windowControllable reasoningZero data retentionNo training on inputs (attested)
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, include_reasoning, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p