Abliteration

Abliterated Large Flash

Abliteration's lower-cost large model pairs a million-token context with adjustable reasoning for long documents and multi-step tasks.

Large FlashAdjustable reasoningtext inputtext output

Run this model

abliteration/abliterated-model-large-flash

Use an endpoint supported by this model.

Get a key for this setup →
Thinking guide →
const response = await fetch('https://api.minirouter.sh/v1/chat/completions', {  method: 'POST',  headers: {    Authorization: `Bearer ${process.env.MINIROUTER_KEY}`,    'Content-Type': 'application/json',  },  body: JSON.stringify({    model: 'abliteration/abliterated-model-large-flash',    messages: [{ role: 'user', content: 'Why is the sky blue?' }],    max_tokens: 1024,  }),}) const data = await response.json()console.log(data.choices[0].message.content)
Read docs →

Start with text and an explicit output limit. Allow room for reasoning as well as the final answer; check the route details for supported input types and limits.

Model overview

Input / 1M
$1.05
All-in MiniRouter rate
Output / 1M
$2.1
All-in MiniRouter rate
Context
1M
1,048,576 tokens
Max output
33K
tokens
Default reasoning
High
Publisher default
Reasoning off
Supported
Unlike Large v2
Released
20 Oct 2026
publisher release

Compare text-model alternatives or estimate token costs for your workload.

Control the work per answer

Choose how much reasoning a task needs.

Large Flash defaults to high reasoning. The publisher also supports low and max effort, or none to switch reasoning off. Hiding the trace leaves reasoning enabled and still consumes tokens.
  1. [ 01 ]

    Routine work

    Try reasoning off for extraction, formatting and straightforward transformations. Check the result against your task's acceptance criteria.

  2. [ 02 ]

    Multi-step tasks

    Start at high for analysis and debugging. Compare low or max on the same examples to judge whether extra reasoning improves the result.

  3. [ 03 ]

    Measure the whole response

    Compare correctness, elapsed time and total output tokens, including reasoning. A shorter visible answer does not necessarily cost less.

Large Flash and Large v2

Lower token rates, a different reasoning choice.

Abliteration lists Flash below Large v2 on both input and output pricing. Flash can also turn reasoning off. Those differences make it worth evaluating for repeated text tasks; they do not establish a quality or latency advantage on your workload.

Publisher list rates · USD per 1M tokens

Flash input / output
$1 / $2
Large v2 input / output
$3 / $5
Abliteration pricing ↗ and reasoning controls ↗, checked October 23, 2026. Publisher rates; current MiniRouter prices appear above.
View chart data
Publisher rates and reasoning controls
ModelInput / 1MOutput / 1MReasoning off
Large Flash$1$2Yes
Large v2$3$5No
Route

One route, standard clients.

Authenticate with a MiniRouter bearer key. Caller-provided upstream credentials are not used on this route.
Base URLhttps://api.minirouter.sh/v1
Model IDabliteration/abliterated-model-large-flash
Endpoints/v1/chat/completions/v1/messages/v1/responses
Parametersmax_tokens, temperature, top_p, frequency_penalty, presence_penalty, stop, seed, tools, tool_choice, reasoning
Structured outputNot enforced; a requested format is dropped and the answer is plain text.
ModalitiesInput: text · Output: text

Configured providers. Priority follows the live catalog order.

ProviderPriorityInput / 1MOutput / 1MParametersAccess
abliteration
103$1.05$2.1max_tokens, temperature, top_p, frequency_penalty, presence_penalty, stop, seed, tools, tool_choice, reasoningMiniRouter balance

Every model, one key.