MagmaRouter vs ArliAI
ArliAI is a zero-log subscription API with unlimited tokens and requests on paid plans, hosting many roleplay finetunes and Derestricted models. MagmaRouter is pay as you go, sells a short line of uncensored tool calling models, and adds an Anthropic-compatible endpoint and per-key spend caps.
Facts checked October 2026 against ArliAI’s own site, docs and policies.
| MagmaRouter | ArliAI | |
|---|---|---|
| What it is | Hosted gateway for one uncensored, tool capable model line (Xploded). | A subscription inference API for text and image generation, hosting many open models and finetunes (many as LoRAs, mostly 8-bit quantized) behind OpenAI and SDNext compatible endpoints. |
| Pricing | Pay as you go, $5 minimum top-up, no subscription required. | Monthly plans with unlimited tokens and requests on paid tiers: Free $0 (12K context, delayed responses), Starter $10 (models up to 31B, 16K context), Core $15 (up to 355B, 32K), Plus $20 (128K), Pro $30 (256K, high priority), Max $160 (512K). No per-token billing. |
| OpenAI-compatible | Yes | Yes. Drop-in OpenAI-compatible API at https://api.arliai.com with /v1/chat/completions and /v1/completions. Chat completions accepts tools and tool_choice. |
| Anthropic-compatible | Yes, /v1/messages (verified with Claude Code) | No. The API docs list no /v1/messages endpoint, only OpenAI-style chat completions. |
| Uncensored / abliterated | Five uncensored models, the Xploded line: GLM 5.3, GLM 5.3 Flash, DeepSeek V4.1 Flash, MiMo V2.6 Flash and Qwen3.8 27B. Every one calls tools. | Their own roleplay finetune Llama-3.3-70B-ArliAI-RPMax-v3, plus Derestricted builds such as Qwen3.5-27B-Writer-Derestricted and GLM-4.6-Derestricted. |
| Prompt retention | No prompt or completion stored; model, tokens and cost kept 90 days; signed prompt-hash receipt | Privacy policy: they do not log, store or have access to prompts or outputs; content is processed transiently and immediately discarded. Usage metadata (request counts, models used) is kept up to 60 days. Training use is not explicitly addressed. |
| Per-key spend caps | Lifetime and monthly per-key caps, enforced by reserving the worst case and clamping max_tokens | Parallel requests per plan: 1 (Free, Starter, Core, Plus), 2 (Pro), 6 (Max); extra parallel requests are blocked. Free accounts get 5 requests per model every 2 days. Per-key spend limits not documented; keys support parameter overrides. |
Where ArliAI is the better choice
- Unlimited tokens and requests for a flat $10 to $30 a month on most plans, which for heavy roleplay or long chats is far cheaper than paying per token at MagmaRouter's rates.
- A large catalogue of roleplay and Derestricted finetunes, context up to 512K on the top plan, image generation included, and a free tier.
Where MagmaRouter is different
- MagmaRouter has no subscription: pay as you go from a $5 minimum top-up, so occasional users do not pay a monthly fee.
- MagmaRouter exposes an Anthropic-compatible /v1/messages endpoint that works with Claude Code, every Xploded model does tool calling for coding agents, and keys get hard lifetime and monthly spend caps with max_tokens clamped before each call.
The uncensored models MagmaRouter sells
If you are comparing ArliAI because you want an uncensored or abliterated model API, these are the models on MagmaRouter: GLM 5.3 uncensored, GLM 5.3 Flash uncensored, DeepSeek V4.1 Flash uncensored, MiMo V2.6 Flash abliterated, Qwen3.8 27B uncensored and Qwen3.8 27B Fast uncensored. Every one calls tools, works in Claude Code through the Anthropic endpoint, and is priced per million tokens as listed on its model page.
| Model | What it is | Context | Price in / out | Supports |
|---|---|---|---|---|
| GLM 5.3 Xploded glm-5.3-xploded | GLM 5.3 uncensored API | 1M | $2.00 / $5.00 | tools, reasoning |
| GLM 5.3 Flash Xploded glm-5.3-flash-xploded | GLM 5.3 Flash uncensored API | 1M | $0.30 / $1.00 | tools, vision, reasoning |
| DeepSeek V4.1 Flash Xploded deepseek-v4.1-flash-xploded | DeepSeek V4.1 Flash uncensored API | 1M | $0.30 / $1.50 | tools, reasoning |
| MiMo V2.6 Flash Xploded mimo-v2.6-flash-xploded | MiMo V2.6 Flash abliterated API | 1M | $1.00 / $2.00 | tools, vision, reasoning |
| Qwen3.8 27B Xploded qwen3.8-27b-xploded | Qwen3.8 27B uncensored API | 1M | $1.50 / $5.00 | tools, vision, reasoning |
| Qwen3.8 27B Fast Xploded qwen3.8-27b-fast-xploded | Qwen3.8 27B Fast uncensored API | 524K | $1.50 / $5.00 | tools, vision, reasoning |
FAQ
Is ArliAI cheaper than MagmaRouter?
For heavy use, yes. ArliAI's paid plans include unlimited tokens for a flat monthly price, while MagmaRouter bills every token at its own per-token prices. MagmaRouter only comes out ahead if your usage is light enough that a few dollars of credit beats a monthly fee.
Does ArliAI work with Claude Code?
ArliAI documents no Anthropic /v1/messages endpoint, so Claude Code cannot point at it directly, though its OpenAI-style chat completions does accept tools. MagmaRouter offers an Anthropic-compatible endpoint built for this.
Does ArliAI keep logs?
ArliAI states it does not log, store or have access to prompts or outputs, keeping only usage metadata for up to 60 days. Its policy does not explicitly address training.
Which uncensored models does MagmaRouter sell?
MagmaRouter sells GLM 5.3 Xploded, GLM 5.3 Flash Xploded, DeepSeek V4.1 Flash Xploded, MiMo V2.6 Flash Xploded, Qwen3.8 27B Xploded, Qwen3.8 27B Fast Xploded. The table above has the price and context for each.
Are these models abliterated or uncensored?
MiMo V2.6 Flash Xploded is abliterated: the refusal direction was removed from the weights. GLM 5.3 Xploded, GLM 5.3 Flash Xploded, DeepSeek V4.1 Flash Xploded, Qwen3.8 27B Xploded, Qwen3.8 27B Fast Xploded are uncensored: the refusal was trained back out. The abliterated vs uncensored page explains the difference.
Which one answers fastest?
MiMo V2.6 Flash Xploded usually starts answering in one to two seconds. The GLM 5.3 Flash model thinks before it answers, so its first word can take much longer.