Qwen3 Coder Flash is available through the abliteratly chat-completions endpoint. Use the model ID below, keep your OpenAI-compatible client, and pay per token from a prepaid crypto balance.
This model is available through the API but is not in the featured index. The page is kept out of search results to keep the directory focused; the model directory lists the featured models.
| Item | USD per 1M tokens |
|---|---|
| Uncached input | $0.195 |
| Cached input | $0.039 |
| Output | $0.975 |
A request with 100,000 input tokens and 20,000 output tokens costs about $0.0390 at the current rates, before any cached-input discount. The pricing page has a calculator and the billing rules.
curl https://abliteratly.beer/v1/chat/completions \ -H "Authorization: Bearer $ABLITERATLY_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen3-coder-flash", "messages": [{"role": "user", "content": "Hello"}] }'
Python:
from openai import OpenAI
client = OpenAI(base_url="https://abliteratly.beer/v1", api_key="sk-ub-...")
r = client.chat.completions.create(
model="qwen3-coder-flash",
messages=[{"role": "user", "content": "Hello"}]
)
print(r.choices[0].message.content)
This is a frontier model. It is the better choice when you need strong instruction following, coding or analysis and do not need uncensored behaviour. It is passed through at the upstream list price; the margin is on the uncensored tier.
| Model | Type | Context | Input | Output |
|---|---|---|---|---|
| Claude Fable 5 claude-fable-5 | frontier | 1.00M | $10 | $50 |
| Claude Fable 5.1 claude-fable-5.1 | frontier | 1.00M | $10 | $50 |
| Claude Haiku 4.5 claude-haiku-4.5 | frontier | 200.0K | $1 | $5 |
| Claude Haiku 5.5 claude-haiku-5.5 | frontier | 1.00M | $0.1 | $0.5 |
Compare the full frontier catalog, check the quickstart, or read the abliterated models guide.
$0.195 per 1M input tokens and $0.975 per 1M output tokens. Cached input is $0.039 per 1M where the provider supports prompt caching.
No. It is a frontier model, not an uncensored or abliterated model. If you need fewer refusals, browse the uncensored tier.
Yes, the model advertises tool support. Test a multi-turn tool loop before relying on it in production.
Set base_url to https://abliteratly.beer/v1, use your API key, and set model to qwen3-coder-flash. The endpoint is POST /v1/chat/completions.
Prompt and response bodies are not written to our database. We store request metadata such as model, token counts, cost and status. The upstream provider's retention policy applies to the request itself.