deepseek/deepseek-v4.1-flashnew model
deepseek v4.1 flash joins the gateway.
$ impossibl changelog
last shipped
openai/gpt-image-2.5-flarenew model
openai/gpt-image-2.5-flare is live on /v1/images/generations at $5.00 input and $30.00 output per 1M tokens.
openai/gpt-image-2.5-sunburstnew model
openai/gpt-image-2.5-sunburst is live on /v1/images/generations at $5.00 input and $30.00 output per 1M tokens.
openai/gpt-6-astranew model
gpt-6 astra joins the gateway with a 1,050,000 token context window and text plus image input.
meta/muse-spark-1.3new model
muse spark 1.3 and muse spark 1.3 contributor join the ai api.
workspacesnew feature
workspaces join the console: invite your team, give each member a role — owner, admin, developer, or viewer — and switch between the workspaces you belong to from the top of the console. keys, balance, and usage belong to the workspace.
google/gemini-3.8-flashnew model
gemini 3.8 flash joins the ai api with a 1m-token context window. it is promo-priced at $0.75 / $3.75 per million tokens through 2026-12-31.
anthropic/claude-fable-5-1new model
claude fable 5.1 joins the ai api with a 1m-token context window and adaptive thinking. cache reads bill at $0.25 per million tokens, a quarter of fable 5's rate.
zai/glm-5.3-flashnew model
glm 5.3 flash joins the ai api.
qwen/qwen3.8-27bnew model
qwen3.8 27b joins the ai api.
zai/glm-5.3new model
glm 5.3 joins the ai api.
deepseek/deepseek-v4-pro-0813new model
deepseek v4 pro 0813 joins the ai api.
google/gemini-3.7-flashnew model
gemini 3.7 flash joins the gateway.
xai/grok-4.6new model
grok 4.6 joins the gateway.
qwen/qwen3.8-2.4t-a95bnew model
qwen3.8 2.4t a95b joins the ai api.
meta/muse-glimmer-30bnew model
muse glimmer 30b joins the ai api.
nvidia/nemotron-3.5-lightningnew model
nemotron 3.5 lightning joins the ai api.
log exports
take your request log with you. filter the logs room, press export csv, and the whole filtered window walks itself into a spreadsheet — not just the rows on screen.
sub-cent costs
a cheap request no longer reads as free. costs under a cent keep their digits down to the credit, so a log of small calls stops printing as a column of $0.00.
qwen/qwen3.8-maxnew model
qwen3.8 max joins the ai api.
deepseek/deepseek-v4-flash-0731new model
deepseek v4 flash 0731 joins the ai api.
google/gemini-3.1-flash-tts-previewnew model
gemini 3.1 flash tts brings 30 steerable voices to /v1/audio/speech, billed per token like every other gemini model.
google/gemini-robotics-er-2-previewnew model
gemini robotics-er 2 preview is now available.
xai/grok-voice-tts-1.0new model
grok voice tts brings five expressive voices and mp3, wav or pcm output to the new /v1/audio/speech endpoint.
audio and video input
send audio and video to gemini models on chat completions, the responses api, and the gemini-native endpoint, with audio billed at the provider's audio rate.
openai/gpt-transcribenew model
gpt-transcribe brings high-accuracy speech-to-text for completed files and streamed transcripts to /v1/audio/transcriptions.
embeddingsnew model
the ai api now turns text into vectors on /v1/embeddings, starting with openai/text-embedding-3-small and openai/text-embedding-3-large.
image generationnew model
the ai api now generates images, starting with openai/gpt-image-2 on /v1/images/generations.
anthropic/claude-opus-5new model
claude opus 5 joins the ai api with a 1m-token context window and native anthropic fallback.
tencent/hy3new model
hunyuan 3 joins the ai api through atlascloud.
qwen/qwen3.7-plusnew model
qwen 3.7 plus joins the ai api through fireworks.
atlascloudnew provider
atlascloud joins the gateway, serving mimo 2.5 and minimax m3.
google/gemini-3.5-flash-litenew model
gemini 3.5 flash-lite is now available.
google/gemini-3.6-flashnew model
gemini 3.6 flash is now available.
logs
every request is now logged. browse them in the dashboard, or have agents pull their own from /v1/requests.
xiaomi/mimo-v2.5new model
mimo 2.5 joins the ai api through atlascloud.
minimaxai/minimax-m3new model
minimax m3 joins the ai api through atlascloud.
realtime dashboard
usage and balance update live in the dashboard as requests come in.
xai/grok-4.5new model
grok 4.5 joins the gateway.
qwennew provider
alibaba's qwen models join the gateway.
qwen/qwen3.8-max-previewnew model
qwen's 3.8 max preview joins the gateway.
faster failovers
failed routes no longer slow down every request. in our benchmarks, failover times dropped from 6 seconds to under 1 millisecond.
auto top-up
recharge when your balance runs low.
moonshotai/kimi-k3new provider
moonshot joins the gateway.
/onboarding
a better start.
thinkingmachines/inklingnew model
two context tiers: 64k and 256k.
cerebras, groq, fireworksnew provider
three fast-inference providers join the gateway.
google gen ai sdk
one-liner migration now supported
openai/gpt-5.6-lunanew model
luna joins the gateway.
openai/gpt-5.6-solnew model
sol joins the gateway.
openai/gpt-5.6-terranew model
terra joins the gateway.
web_search on /v1/responses
xai's native web_search and x_search tools.
deepseek v4 + v3.2new model
v4 and v3.2 join the gateway.
byok
bring your own provider keys now supported
anthropic/claude-fable-5new model
fable 5 redeployed.
accounts + provider pricing
humans can claim agent-created accounts and merge balances. usage stays at provider price.
anthropic/claude-sonnet-5new model
sonnet 5 launches.
z.ainew provider
glm models join the gateway.
human dashboard
api keys, credit top-ups and the model catalog.
/v1/messages, images & tools
anthropic messages support, with image input and tool calling on every surface.
hello world
impossibl init