Updated

Abliterated model API ids

These are the ids an abliterated model API takes in its model field, here on Silk Compute. Every id has a 1,048,576-token context window, tool calling and streaming.

Ids and limits

Kimi K3.1 Pro$3.50 / $8.00 per 1MModel pageDeepSeek V4.1 Flash 1017$0.24 / $0.90 per 1MModel pageDeepSeek V4.1 Flash 1017 Fast$0.24 / $0.90 per 1MModel pageGLM-5.3$2.00 / $4.00 per 1MModel pageGLM-5.3 Flash$0.24 / $1.00 per 1MModel page
Idkimi-k3.1-pro-uncensoreddeepseek-v4.1-flash-1017-uncensoreddeepseek-v4.1-flash-1017-fast-uncensoredglm-5.3-uncensoredglm-5.3-flash-uncensored
MakerMoonshot AIDeepSeekDeepSeekZ.aiZ.ai
Size2,780B763B763B753B320B total, 18B active
Inputtext, images, videotext, imagestext, imagestexttext, images
Context1,048,5761,048,5761,048,5761,048,5761,048,576
Max output943,718131,072131,072131,072131,072
Default effortmaxhighhighhighhigh
Use it forSecurity research, red teaming, agentic codingBest valueSame model, own cached rateSoftware work, text onlyVolume work

Retired ids return 404 model_not_found. Prices are on the pricing page.

List them from the API

GET /models returns the active ids in OpenAI's list format. A model list back also proves the key works.

https://api.silkcompute.com
curl https://api.silkcompute.com/models \
  -H "Authorization: Bearer $SILK_API_KEY"

Example response

{
  "object": "list",
  "data": [
    {"id": "kimi-k3.1-pro-uncensored", "object": "model"},
    {"id": "deepseek-v4.1-flash-1017-uncensored", "object": "model"},
    {"id": "deepseek-v4.1-flash-1017-fast-uncensored", "object": "model"},
    {"id": "glm-5.3-uncensored", "object": "model"},
    {"id": "glm-5.3-flash-uncensored", "object": "model"}
  ]
}
for model in client.models.list():
    print(model.id)

Example response

{
  "object": "list",
  "data": [
    {"id": "kimi-k3.1-pro-uncensored", "object": "model"},
    {"id": "deepseek-v4.1-flash-1017-uncensored", "object": "model"},
    {"id": "deepseek-v4.1-flash-1017-fast-uncensored", "object": "model"},
    {"id": "glm-5.3-uncensored", "object": "model"},
    {"id": "glm-5.3-flash-uncensored", "object": "model"}
  ]
}
for await (const model of client.models.list()) {
  console.log(model.id)
}

Example response

{
  "object": "list",
  "data": [
    {"id": "kimi-k3.1-pro-uncensored", "object": "model"},
    {"id": "deepseek-v4.1-flash-1017-uncensored", "object": "model"},
    {"id": "deepseek-v4.1-flash-1017-fast-uncensored", "object": "model"},
    {"id": "glm-5.3-uncensored", "object": "model"},
    {"id": "glm-5.3-flash-uncensored", "object": "model"}
  ]
}