DevPass

One AI coding subscription. Every model.
by @llmgateway

[DevPass @LLMDevPass]

Biggest 500M+ faller: GLM-5.3, -45.9% to 817.6M tokens, now #7.

See if Sonnet overtakes MiniMax, with all 132 models ranked live: https://t.co/WEzaMgwsh4

[DevPass @LLMDevPass]

A photo finish for the last spot on the chart.

#8 MiniMax M3.1 Flash: 710.2M tokens, +675.5%
#9 Claude Sonnet 5.5: 707.1M, +104.8%

3.1M tokens apart, out of 33.5B through LLM Gateway in 24h (1.2M requests). DeepSeek V4.1 Flash 1017 led with 44.8%, down 9.9%.

[DevPass @LLMDevPass]

MiniMax H3 Image is live on LLM Gateway: MiniMax's image generation and editing model. Prompt enhancement, up to 16 reference images, output up to 4K. $2.30/M text in, $3.80/M image out. https://t.co/mKgSKGkzA4

[DevPass @LLMDevPass]

Gemini 4 Hydrogen is live on LLM Gateway: Google's fast, efficient model for everyday coding and agentic workflows. 1M context, vision + audio + tools + reasoning up to xhigh. $1.50/M in, $7.50/M out. https://t.co/ozGCPx2Vbu

[DevPass @LLMDevPass]

Scribe v3 Live is live on LLM Gateway: ElevenLabs' realtime speech-to-text with turn prediction and live diarization across 90+ languages. Streams over /v1/realtime. $0.22 per hour of audio.
https://t.co/Ia1ALfMKi4

[DevPass @LLMDevPass]

Unpopular opinion: switching coding agents shouldn't mean switching subscriptions.

DevPass: one key in Claude Code, Cursor, Cline or any OpenAI/Anthropic-compatible agent. $29/mo → $58 of usage at provider list rates, up to 8% of it per rolling 24h.

https://t.co/zfkVQH6L8K

[DevPass @LLMDevPass]

Jev 1.14 averaged ~6.2k tokens a call vs 24.7k board-wide. Biggest 500M+ faller: GPT-6 Luna, -39.1%. All 120 models, ranked live: https://t.co/WEzaMgwsh4

[DevPass @LLMDevPass]

7 of the 8 charted models on LLM Gateway grew in the last 24h. The one that shrank is #1.

DeepSeek V4.1 Flash 1017: -30.9% to 16.9B tokens, still 40.6% of 41.5B.

Fastest climber: Jev 1.14, +773.6% to 1.0B, now #8. MiMo V2.6 Pro: +152.6% to 3.2B, #4.

[DevPass @LLMDevPass]

Scribe v3 is live on LLM Gateway: ElevenLabs' most accurate speech-to-text model, with word-level timestamps and speaker diarization across 90+ languages. $0.20 per hour of audio. https://t.co/rTKtVBp3o7

[DevPass @LLMDevPass]

Unused DevPass allowance now carries over.

At renewal, what's left joins your next cycle, up to half the plan's monthly allowance: $29 on Lite, $79 on Pro, $179 on Max. Nothing to switch on, and the daily and weekly premium caps rise with it.

https://t.co/ZaegtyCc72

[DevPass @LLMDevPass]

Four fast text-to-image models are live on LLM Gateway, served via fal: FLUX 3 Klein, FLUX.2 Klein 4B, Z-Image Turbo and Recraft V4.1 Flash. From $0.005/image at 1024x1024. https://t.co/bnXeTChgXd

[DevPass @LLMDevPass]

MiniMax H3 is live on LLM Gateway: MiniMax's flagship video model with native synchronized audio. Text-to-video and first/last-frame image-to-video, 4-15s clips at 768p or 2K. $0.08/sec at 768p, $0.13/sec at 2K. https://t.co/3UZtRvIG7C

[DevPass @LLMDevPass]

Biggest 500M+ faller: GPT-6 Luna, -50.8% to 1.5B tokens. Hour by hour, all 173 models: https://t.co/WEzaMgwsh4

[DevPass @LLMDevPass]

3,639 requests. 1.3B tokens.

MiMo V2.6 Pro grew 417.8% on LLM Gateway in 24h, to #7, at ~347.5k tokens per call vs 30.3k board-wide.

GLM-5.3 Flash also more than doubled: +130.9% to 4.2B, #3.

46.8B tokens, 1.5M requests. DeepSeek V4.1 Flash 1017 took 52.3%.

[DevPass @LLMDevPass]

Orpheus V1 English and Orpheus Arabic Saudi are live on LLM Gateway: Canopy Labs' expressive text-to-speech, served via Groq. Six voices each, WAV output. $22/1M characters for English, $40/1M for Saudi Arabic. https://t.co/dYWBdiggn5

[DevPass @LLMDevPass]

KLING 3.0 Omni is live on LLM Gateway: Kling Video O3 via AtlasCloud. Text-to-video, first/last-frame and reference image or video inputs, 3-15s clips, 720p to 4K, optional audio. From $0.084/sec at 720p, $0.42/sec at 4K. https://t.co/cEKhAwCyTY

[DevPass @LLMDevPass]

미리 알아 두실 점입니다. 요금은 USD로 청구되고 원화 결제는 지원하지 않습니다. 일일 사용 한도(연속된 24시간 기준 월 한도의 8~10%)가 있습니다. 한도의 10% 미만을 썼다면 첫 결제 후 14일 안에 대시보드에서 직접 환불할 수 있습니다. 한국어 표현이 어색한 곳이 있으면 알려 주세요.

[DevPass @LLMDevPass]

DevPass 한국어 페이지를 열었습니다.

AI 코딩 구독 하나로 Claude, GPT, Gemini를 지금 쓰는 에이전트에서 그대로.
- 월 $29부터, 결제 금액의 2배만큼 사용
- 남은 한도는 월 한도의 50%까지 이월
- 약정 없음

https://t.co/e6KOYNKCst

[DevPass @LLMDevPass]

Luma Ray 3.2 is live on LLM Gateway: cinematic video, served via AtlasCloud. Text-to-video (5 or 10s) and first/last-frame image-to-video at 540p, 720p or 1080p, no audio. From $0.033/sec at 540p, $0.066/sec at 720p. https://t.co/Oc0cPuQ0ab

[DevPass @LLMDevPass]

Gemini Omni 1.2 Flash is live on LLM Gateway: Google's video model with native audio. Generation and editing, 3-15s clips, 360p to 4K. From $0.023/sec at 360p, $0.07/sec at 720p. https://t.co/paIjkyq1PA