Qwen3 8B
novita-ai/qwen/qwen3-8b-fp8 · in: text → out: text · price provider-declared (via models.dev). Capabilities: reasoning text.
$0.06
Blended $/1M (3:1)
$0.035
Input $/1M
$0.138
Output $/1M
$0.17
1M in + 1M out
—
Cache read $/1M
#744 of 7,712
Cheapest rank
- Provider
- NovitaAI (
novita-ai) - Kind
- text (outputs text)
- Context
- 128,000 tokens
- Max output
- 20,000 tokens
- Capabilities
- reasoning text
Same model, other providers
“Qwen3 8B” is offered by 5 providers — cheapest-effective first. This one (NovitaAI) is the cheapest.
| Provider | Input | Output | Blended | Context |
|---|---|---|---|---|
| NovitaAI (this) | $0.035 | $0.138 | $0.06 | 128k |
| Alibaba (China) | $0.072 | $0.287 | $0.13 | 131k |
| Pioneer | $0.2 | $0.2 | $0.20 | 41k |
| OpenRouter | $0.117 | $0.455 | $0.20 | 131k |
| Alibaba | $0.18 | $0.7 | $0.31 | 131k |
Call it
jishie indexes & prices models; it does not proxy inference.
Most providers are OpenAI-compatible — point base_url at NovitaAI and pass this model id:
curl $BASE_URL/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
-d '{"model":"qwen/qwen3-8b-fp8","messages":[{"role":"user","content":"hello"}]}'
Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):