Naar inhoud
Draait in:FranceGemaakt in:China
OVH AI Endpoints (GRA)

Qwen3.5-397B-A17B

Tokonomix-redactie·Gecontroleerd door Mes Kalkan··
Sectie 01

Snelheidsanalyse

Latency gemeten over alle benchmark-runs. P50 (mediaan) en P95 (95e percentiel) geven een realistisch beeld van de responssnelheid onder normale en piekbelasting.

P50 latency (mediaan)P95 latency53 runs
16788715758236293150005-2806-10ms
Sectie 02

Kwaliteitsscores

Evaluatieresultaten van judge-model beoordelingen over diverse taakcategorieën. Scores weerspiegelen coherentie, accuratesse en instructieopvolging.

100
Code generatie
45
Creatief
1
Feitelijk
30
Meertaligheid
Sectie 03

Prijsgeschiedenis

Directe provider-tarieven per miljoen tokens, plus een typische gespreks-kostschatting.

💰
API-tarieven — Qwen3.5-397B-A17B
$0.7100 per 1M input-tokens
$4.25 per 1M output-tokens
≈ $0.0013 per typisch gesprek (800 tokens)
Input vs output prijs (per 1M tokens)
per 1M input-tokens$0.7100
per 1M output-tokens$4.25
No pricing history yet — will populate after the first metadata sync detects a price change.
Sectie 04

Tokens per seconde

Doorvoersnelheid in tokens per seconde, afgeleid uit gemeten P50-latency. Hogere waarden zijn beter; fluctuaties weerspiegelen serverbelasting bij de provider.

Doorvoer (tokens / s)760 / avg 1195
122235

Geschat uit P50-latency × 200 output-tokens — het absolute getal hangt af van deze aanname; de trend is wat telt.

Sectie 05

Mogelijkheden

ownedBy: Qwen
Sectie 06

Tokonomix benchmark-oordelen

⚖️
Endorsed by 1 judge
Independent LLM judges evaluated this model on our weekly intelligence tests
claude-sonnet-4-535/100 · 7 runs
2 correct0 partial5 wrong29% accuracy
2026-05-31

Qwen3.5-397B-A17B establishes baseline with strong creative performance

This first benchmark window establishes baseline performance for Qwen3.5-397B-A17B deployed through OVH AI Endpoints in the GRA region. The model demonstrates particularly strong creative writing capabilities, achieving 9.0 out of 10 in creative tasks, indicating robust narrative generation and imaginative content production. Coding performance is solid at 7.5, showing competence in programming tasks though with room for optimization. Mathematical reasoning scores 7.0, representing adequate performance for standard computational problems. The model handles instruction following reliably at 7.0, meeting basic compliance requirements. Response coherence is maintained at 7.0, ensuring outputs remain logical and well-structured. Overall performance across all categories averages a respectable level for a model of this class. Users should expect best results when leveraging the model for creative content generation, storytelling, and narrative tasks. For production code generation and complex mathematical proofs, outputs may require additional validation. This baseline provides a reference point for tracking future performance trends and model updates.

Quality

Latency p50

Test runs

0

Strong creative writing at 9.0 Solid coding performance at 7.5 Math reasoning needs improvement Baseline established across all metrics
Laatste automatische test
10 jun 2026 · 02:00 UTC · Snelheidstest
P50 latency
263 ms
P95 latency
279 ms
Fouten
0 / 6 runs
Laatst beoordeeld door Tokonomix-team·10 juni 2026