Vercel AI Gateway — 模型目录与真实网关流量排行榜

Vercel AI Gateway

模型目录 · 能力/定价/性能对比 + 真实网关流量排行

目录 476 个模型 · 更新于 2026-10-19
模型
上下文
延迟
吞吐
输入价
输出价
缓存读
联网搜索
能力
deepseek-v4.1-flash-1017
DeepSeek· 2026-10-17
1.0M 411 ms 464 tps $0.07 $0.30 $0.00 —
+2
deepseek-v4.1-flash-1017-fast
DeepSeek· 2026-10-17
1.0M 696 ms 745 tps $0.60 $2.40 $0.01 —
+2
glimpse
elorian· 2026-10-16
220K — — $0.60 $3.00 $0.06 —
pplx-decider-flash-v1.2
Perplexity· 2026-10-16
262K — — $0.01 $0.00 — —
pplx-decider-v1.2
Perplexity· 2026-10-16
262K — — $0.02 $0.00 — —
beam
reflection· 2026-10-16
1.0M 146 ms 392 tps $0.00 $0.00 — —
+1
beam-free
reflection· 2026-10-16
1.0M — — $0.00 $0.00 — —
+1
flux-3-image-klein
Black Forest Labs· 2026-10-15
— — — — — — —
d1-ultra
liquid· 2026-10-15
33K — — $0.04 $0.00 — —
muse-spark-open
Meta· 2026-10-15
1.0M 350 ms 279 tps $1.25 $4.25 $0.15 —
+3
nemotron-3.5-super-vl
英伟达· 2026-10-15
1.0M 263 ms 719 tps $0.20 $0.50 $0.04 —
+2
cat-jam
stealth· 2026-10-15
1.0M 443 ms 369 tps $0.00 $0.00 — —
jev-1.14.0
typesafe-ai· 2026-10-15
32K — — $0.04 $0.00 — —
nemotron-3.5-lightning-free
英伟达· 2026-10-14
1M 418 ms 292 tps $0.00 $0.00 — —
+1
drama-3
Fish Audio· 2026-10-13
— — — $10.00 — — —
gemini-4-argon
Google· 2026-10-13
1M 4.1 s 73 tps $2.00 $10.00 $0.10 $14.00
+4
lfm2.5-24b-a2b
liquid· 2026-10-13
128K 193 ms 244 tps $0.05 $0.70 $0.03 —
mai-image-2.6
微软· 2026-10-13
— — — $5.00 $38.00 — —
mai-image-2.6-flash
微软· 2026-10-13
— — — $1.75 $19.00 — —
scaledown-classify-1
scaledown· 2026-10-13
— — — $0.05 $0.00 — —
muse-wonder-1.0
Meta· 2026-10-12
1.0M 3.2 s 110 tps $4.00 $20.00 $0.20 —
+2
heygen-video-1
heygen· 2026-10-09
— — — — — — —
microsoft-decision-1
微软· 2026-10-09
33K — — $0.04 $0.00 — —
apex
callstack· 2026-10-08
262K 1.0 s 138 tps $0.50 $3.00 $0.20 —
claude-haiku-5.5
Anthropic· 2026-10-07
1M 834 ms 228 tps $0.10 $0.50 $0.01 $10.00
+3
claude-haiku-5.5-batch
Anthropic· 2026-10-07
1M — — $0.05 $0.25 $0.01 —
+2
glyph-cluster
stealth· 2026-10-07
256K 4.4 s 41 tps $0.00 $0.00 — —
gemini-nano-banana-2.1
Google· 2026-10-06
131K 775 ms 129 tps $1.50 $7.50 $0.15 $14.00
+2
mistral-large-4
Mistral· 2026-10-06
524K 1.8 s 81 tps $0.68 $2.09 $0.07 —
gpt-6-luna-decisions
OpenAI· 2026-10-06
1.1M — — $0.10 — — —
proteus
topaz· 2026-10-06
— — — — — — —
starlight-precise-2.6
topaz· 2026-10-06
— — — — — — —
wonder-3.5
topaz· 2026-10-06
— — — — — — —
flux-3-image
Black Forest Labs· 2026-10-05
— — — — — — —
laya
convaiinnovations· 2026-10-01
8K — — $0.00 $0.00 — —
laya-free
convaiinnovations· 2026-10-01
8K — — $0.00 $0.00 — —
mai-transcribe-2-streaming
微软· 2026-10-01
— — — — — — —
mai-voice-2.1
微软· 2026-10-01
— — — $22.00 — — —
mai-voice-2.1-flash
微软· 2026-10-01
— — — $15.00 — — —
grok-imagine-video-1.5-lite
xAI· 2026-10-01
— — — — — — —
embed-v5.0-fast
Cohere· 2026-09-30
128K — — $0.08 — — —
embed-v5.0-pro
Cohere· 2026-09-30
128K — — $0.12 — — —
rerank-3
Voyage· 2026-09-30
32K — — $0.05 — — —
rerank-3-lite
Voyage· 2026-09-30
32K — — $0.02 — — —
deepseek-v4.1-flash-fast
DeepSeek· 2026-09-29
1.0M 1.1 s 652 tps $0.60 $2.40 $0.01 —
+2
ling-3.1-flash
InclusionAI(蚂蚁)· 2026-09-29
262K 955 ms 102 tps $0.30 $0.90 $0.06 —
d1
liquid· 2026-09-29
66K — — $0.02 $0.00 — —
gpt-6.1-sol
OpenAI· 2026-09-29
1.1M 1.7 s 99 tps $2.00 $10.00 $0.10 $10.00
+5
gpt-6.1-sol-fast
OpenAI· 2026-09-29
1.1M 1.2 s 102 tps $4.00 $20.00 $0.20 $10.00
+5
gpt-6.1-sol-batch
OpenAI· 2026-09-29
1.1M — — $1.00 $5.00 $0.05 —
+3
claude-sonnet-5.5
Anthropic· 2026-09-28
1M 1.1 s 153 tps $2.00 $10.00 $0.10 $10.00
+3
claude-sonnet-5.5-batch
Anthropic· 2026-09-28
1M — — $1.00 $5.00 $0.05 —
+2
longcat-2.5-preview
meituan· 2026-09-26
1.0M 2.3 s 71 tps $0.30 $1.20 $0.01 —
+1
qwen3.8-max-prime
阿里巴巴· 2026-09-23
1M 1.0 s 120 tps $4.00 $12.00 $0.50 —
+2
ember-1
fireworks· 2026-09-23
1.0M 642 ms 90 tps $3.00 $15.00 $0.30 —
gemini-3.8-flash-lite-tts
Google· 2026-09-23
— — — $0.50 $6.00 — —
gemini-3.8-flash-tts
Google· 2026-09-23
— — — $0.50 $9.00 — —
recraft-v4.1-flash
Recraft· 2026-09-23
— — — — — — —
claude-opus-5.5
Anthropic· 2026-09-22
1M 1.7 s 133 tps $4.00 $20.00 $0.20 $10.00
+4
claude-opus-5.5-fast
Anthropic· 2026-09-22
1M 1.5 s 268 tps $8.00 $40.00 $0.40 $10.00
+4
共 476 个模型 · 第 1 / 8 页
口径:延迟=p50 首字延迟 · 吞吐=p50 每秒输出 token · 价格均为每百万 token($/M)· 数据来源 vercel.com/ai-gateway · 每 2h 同步