modelbenchmark.io

WizardLM 2 8x22B (Deep Infra)

Microsoft · wizardlm-2-8x22b

Compare

Microsoft WizardLM instruction-tuned Mixtral MoE for multilingual dialogue and complex tasks

Specification

most-agreed values

Context
66K
Max output
8K
Released
2024-04-15
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.48
Output
$0.48
Cache read
$0.2465

Quality

2 benchmarks · 1 source

Composite
20th
percentile of 344 scored models
Rank
276 / 344
±0.06 sd
Evidence
2 × 1
too little evidence to place with confidence
Effort range
—
one configuration only
reasoning20th1/4 bench
math20th1/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 525.7 ±0.9defaultEpoch AI2025-01-27
GPQA diamond43.4 ±2.3defaultEpoch AI2025-01-27

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 5 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
deepinfra$0.48$0.48——66K66K—
edenai$0.48$0.48——66K8K—
nano-gpt$0.493$0.493$0.2465—66K8K—
novita$0.62$0.62——66K8K—
novita-ai$0.62$0.62——66K8K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02openrouterInput$0.62
2026-09-19openrouterInput$0.62
2026-10-10openrouterInput$0.62
2026-09-02openrouterOutput$0.62
2026-09-19openrouterOutput$0.62
2026-10-10openrouterOutput$0.62

More from Microsoft

most-hosted first