modelbenchmark.io

Qwen3.5 397B-A17B

Alibaba · qwen3-5-397b-a17b

Compare

Large open Qwen multimodal MoE for visual agents and long technical tasks

Specification

alibaba listing

Context
262K
Max output
66K
Released
2026-02-15
Knowledge cutoff
2025-07
Retires
2026-06-29
Open weights
yes
Input
text, image, video
Output
text

Price

US dollars per million tokens · alibaba list price

Input
$0.60
Output
$3.60
Cache read
$0.22

Quality

3 benchmarks · 1 source

Composite
74th
percentile of 344 scored models
Rank
92 / 344
±0.07 sd
Evidence
3 × 1
one source, or fewer than 3 benchmarks
Effort range
—
one configuration only
reasoning88th1/4 bench
math63rd2/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
OTIS Mock AIME 2024-202588.9 ±4.7defaultEpoch AI2026-08-07
GPQA diamond85.9 ±2.5defaultEpoch AI2026-08-07
FrontierMath-Tiers-1-3-v229.5 ±2.7defaultEpoch AI2026-08-28

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

Available from 39 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
alibabavendor$0.60$3.60——262K66K—
unorouter · freefreefree——262K66K—
aihubmix$0.1644$0.9864——262K66K—
aihubmix · aihubmix$0.1644$0.9864——262K66K—
alibaba-cn$0.172$1.032——262K66K—
empiriolabs$0.172$1.032$0.172—258K64K—
merge-gateway$0.172$1.032$0.0344—131K33K—
orcarouter$0.172$1.032——262K66K—
qiniu-ai$0.1739$1.0435——256K64K—
kilo$0.39$2.34——262K236K—
nano-gpt$0.39$2.34$0.30—262K66K—
nano-gpt · thinking$0.39$2.34$0.30—262K66K—
siliconflow$0.39$2.34——262K262K—
tokengo$0.40$2.65$0.20—262K66K—
deepinfra$0.45$3.00$0.22—262K
82K
262K
—
nearai$0.50$3.30$0.25—262K66K—
openrouter$0.50$3.60$0.30—262K236K—
digitalocean$0.55$3.50$0.111—131K131K—
nano-gpt · tee$0.55$3.50$0.275—262K66K—
ofox$0.55$3.50$0.55—256K64K—
ofox · bailian$0.55$3.50$0.55—256K64K—
cloudflare-ai-gateway$0.60$3.60——262K66K—
huggingface$0.60$3.60——262K33K—
jalapeno$0.60$3.60——262K66K—
llmtr$0.60$3.60——256K66K—
mixlayer$0.60$3.60——262K262K—
nebius$0.60$3.60$0.12—262K
8K
262K
—
novita$0.60$3.60——262K66K—
novita-ai$0.60$3.60——262K66K—
scaleway · scaleway$0.60$3.60——256K16K—
streamlake · streamlake$0.60$3.60$0.12—258K64K—
together_ai$0.60$3.60$0.35—262K262K2026-06-29
togetherai$0.60$3.60$0.35—262K130K—
cortecs$0.668$4.01——262K262K—
scaleway$0.6729$4.0374——256K16K—
ovhcloud$0.71$4.25——262K262K—
venice$0.75$4.50——128K33K—
greenpt$0.798$4.959——262K16K—
thinkingmachines$1.65$7.50$0.33—66K66K—

Price history

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

DateHostField$/M
2026-09-02nebiusCache read$0.06
2026-10-22nebiusCache read$0.60
2026-10-24nebiusCache read$0.12
2026-09-02together_aiCache read$0.35
2026-09-24together_aiCache read$0.35
2026-09-02nano-gptInput$0.60
2026-10-19nano-gptInput$0.39
2026-09-02nebiusInput$0.60
2026-09-06nebiusInput$0.60
2026-09-02openrouterInput$0.39
2026-09-03openrouterInput$0.55
2026-09-06openrouterInput$0.55
2026-09-07openrouterInput$0.39
2026-09-09openrouterInput$0.55
2026-10-06openrouterInput$0.45
2026-10-07openrouterInput$0.45
2026-10-09openrouterInput$0.55
2026-10-10openrouterInput$0.55
2026-10-14openrouterInput$0.50
2026-10-20openrouterInput$0.39
2026-10-22openrouterInput$0.50
2026-09-02scalewayInput$0.60
2026-10-22scalewayInput$0.6729
2026-09-02together_aiInput$0.60
2026-09-24together_aiInput$0.60
2026-09-02nano-gptOutput$3.60
2026-10-19nano-gptOutput$2.34
2026-09-02nebiusOutput$3.60
2026-09-06nebiusOutput$3.60
2026-09-02openrouterOutput$2.34
2026-09-03openrouterOutput$3.50
2026-09-06openrouterOutput$3.50
2026-09-07openrouterOutput$2.34
2026-09-09openrouterOutput$3.50
2026-10-06openrouterOutput$3.00
2026-10-07openrouterOutput$3.00
2026-10-09openrouterOutput$3.50
2026-10-10openrouterOutput$3.50
2026-10-14openrouterOutput$3.60
2026-10-20openrouterOutput$2.34
2026-10-22openrouterOutput$3.60
2026-09-02scalewayOutput$3.60
2026-10-22scalewayOutput$4.0374
2026-09-02together_aiOutput$3.60
2026-09-24together_aiOutput$3.60
2026-09-03openrouterCache read$0.225
2026-09-09openrouterCache read$0.225
2026-09-19openrouterCache read$0.225
2026-10-06openrouterCache read$0.22
2026-10-07openrouterCache read$0.22
2026-10-09openrouterCache read$0.225
2026-10-10openrouterCache read$0.225
2026-10-14openrouterCache read$0.30
2026-10-22openrouterCache read$0.30
2026-09-15nano-gpt · thinkingInput$0.60
2026-10-19nano-gpt · thinkingInput$0.39
2026-09-15nano-gpt · thinkingOutput$3.60
2026-10-19nano-gpt · thinkingOutput$2.34

Other listings of this model

same model, different key

More from Alibaba

most-hosted first