bytedance

Axion 1.0: How ByteDance Built a Fully Open Agent Foundation Model with Just 18K Training Examples Axion 1.0: How ByteDance Built a Fully Open Agent Foundation Model with Just 18K Training Examples Building capable AI agents has historically required elaborate training pipelines — mid-training phases, direct preference optimization, on-policy distillation, and mountains of proprietary data. Axion 1.0, released on November 2, 2026 by a team from ByteDance, Duke University, the University of Maryland, and the University at Buffalo, challenges that assumption. It achieves competitive performance on hard scientific reasoning and tool-use benchmarks using a three-stage recipe, 18,363 training trajectories, and roughly 1,000 GPU-hours for the 122B variant — and releases everything under Apache 2.0. The Axion 1.0 paper and project page make the case that data quality and task construction matter more than pipeline complexity. Here is what that looks like in practice. The Three-Stage Training Recipe Axion 1.0 deliberately avoids techniques that have become standard in agent training: no dedicated mid-training, no DPO, no on-policy distillation. Instead, it uses two supervised fine-tuning stages followed by reinforcement learning. Stage I — Learn to Reason (SFT): The model starts with self-contained scientific reasoning, meaning tasks that require no external tools. The team uses model-aware data construction: seed problems are synthesized from public sources, then adversarially strengthened based on the base model's own successful solution strategies. Operations like composition, parameterization, and range shifts generate 10,819 trajectories that are genuinely hard for the model at the time of training. Stage II — Learn to Investigate (SFT): The second stage introduces tool-dependent reasoning. Tasks are grounded in real-world databases — NASA Exoplanet Archive, OpenAlex, UniProt — and reference answers are determined programmatically before the LLM formulates the question. This ensures verifiability without human annotation. The stage adds 7,544 trajectories (approximately 312M tokens), teaching the model to recover missing information through external retrieval. Stage III — Learn Through Interaction (RL): The final stage uses asynchronous Group Relative Policy Optimization (GRPO) over long-horizon agent trajectories. Rollout workers interact with tools and environments concurrently with policy updates in the trainer pool, avoiding the idle time that synchronous RL creates. Session-level training handles the long interaction histories that agentic tasks produce. Architecture and Infrastructure Axion 1.0 is available in two sizes: 35B and 122B parameters, both using a Mixture of Experts (MoE) architecture with a 128K context window during SFT. The training infrastructure is built for reproducibility: - FSDP + expert parallelism + context parallelism for SFT runs at 128K context - Token-In-Token-Out (TITO) to preserve token IDs across the pipeline, preventing subtle mismatches between rollout and training - Rollout Routing Replay (R3) to align MoE expert choices between the rollout phase and the training phase — a non-obvious consistency requirement for MoE agents - swift for SFT and verl for RL, providing a unified, documented pipeline The R3 mechanism is worth noting. In a standard MoE setup, the expert routing during rollout (when the model generates trajectories) can differ from routing during the training forward pass, because the two happen in different contexts. R3 replays the routing decisions from rollout into training, ensuring the gradient updates land on the same experts that produced the behavior being reinforced. Benchmark Results Axion 1.0 was evaluated on six benchmarks spanning scientific reasoning, information seeking, and biological data analysis: Humanity's Last Exam (HLE), BioMysteryBench (BMB), SEAL-0, HiPhO, and the Olympiad and Research splits of FrontierScience. | Model | HLE | BMB | SEAL-0 | HiPhO | FS Olympiad | FS Research | |---|---|---|---|---|---|---| | Agents-A1 | 47.6 | — | 56.4 | 46.4 | 79.0 | 40.0 | | Gemini 3.1 Pro | 51.4 | 35.3* | 53.2* | 38.5* | 85.0* | 26.7* | | Axion 1.0-35B | 52.2 | 35.3 | 62.2 | 45.7 | 86.0 | 56.7 | | Axion 1.0-122B | 59.0 | 41.7 | 58.6 | 46.2 | 87.0 | 66.7 | *Evaluated by the Axion team. The 122B model scores 59.0 on HLE and 66.7 on FrontierScience Research — the latter being a particularly demanding benchmark that requires synthesizing information across multiple scientific papers. The 35B model holds its own against Gemini 3.1 Pro on most tasks despite being a fraction of the cost to run. What Makes the Data Construction Interesting The model-aware approach in Stage I is the most technically distinctive part of the recipe. Rather than curating a static dataset, the team generates problems, runs the base model on them, identifies which solution strategies succeed, and then constructs harder variants that stress-test those same strategies. This creates a feedback loop between the model's current capabilities and the difficulty of its training data — without requiring human labelers to judge what "hard" means. Stage II's grounding in real databases is equally deliberate. By anchoring tasks to sources like UniProt and OpenAlex, the team ensures that the model learns to retrieve and reason over evidence that actually exists, rather than hallucinating plausible-sounding citations. The programmatic answer generation also means every training example has a verifiable ground truth, which is what makes GRPO's reward signal reliable in Stage III. What Is Released The GitHub repository contains the full training code: data construction, filtering, SFT, asynchronous RL, and the evaluation harness. The training dataset on Hugging Face provides both SFT splits and the RL tasks. The 35B reproduction checkpoint is publicly available under Apache 2.0. The 122B weights are also released, making this one of the more complete open releases for a model at this performance tier. Practitioner Implications For teams building domain-specific agents, Axion 1.0 offers a few concrete takeaways. First, the three-stage recipe is modular: Stage I can be replaced with domain-specific reasoning data, and Stage II can be grounded in proprietary databases rather than public ones. The infrastructure (TITO, R3, asynchronous GRPO) is reusable. Second, the 18K trajectory count is small enough that fine-tuning on top of the released checkpoints is tractable for teams without large compute budgets. Third, the emphasis on verifiable tasks — where the answer can be checked programmatically — points toward a general principle: RL for agents works best when you can define a reward without human judges in the loop. The broader trend here is that agent training is becoming more systematic. The gap between "we trained an agent" and "here is the exact recipe, data, and code to reproduce it" is closing, and Axion 1.0 is a clear example of what that looks like when done carefully.

Harvesting SWE Benchmarks from Wild Vibe Coding Sessions Ao Lu, Yuanhao Zeng, Shaocheng Lan, Sijie Xu, Jiacheng Bao, Kuang Yan, Heng Lin, Kan Ren arxiv.org [𝚌𝚜.𝚂𝙴 𝚌𝚜.𝙰𝙸] 💬Code: bytedance/SessionFarmer t.co

My PR demo videos make themselves now, in my voice How I built a free pipeline that runs on my MacBook: it films the real app, narrates in a clone of my voice and can put a lip-synced webcam tile of me in the corner. Plus the benchmarks, the bugs, and the 23-minute render that nearly killed the whole idea. !A script draws on the Annotator website while a small webcam tile of me talks in the bottom-left corner. Nothing in this clip was recorded live. A script moves the cursor, the voice is a clone of mine, and the tile in the corner is a two-minute webcam take of me with new lips. Full 76-second video. TL;DR - Every PR in my projects ships with a one-to-three-minute video of the real app, narrated in my voice. I don't record them anymore. - I'm cheap, so it's all free and runs locally on one MacBook Pro: a voice clone (Voicebox), a lip-synced webcam tile (MuseTalk), and a screen recording driven through macOS accessibility APIs. - The avatar used to take 23 minutes to render one minute of video. It now takes about one. - Most of the time went into unglamorous bugs: pauses in the audio, a toolbar that hides itself after three seconds, and my Hebrew keyboard layout. Why I wanted this Most of my PRs are written with a coding agent, usually on autopilot. A demo video in the PR makes it much easier to understand what the author and their agent were going for. Bigger PRs stop being scary, and you can see the before and after for yourself instead of picturing it from the diff. Before AI coding, when you didn't understand someone's change, you'd call them and ask them to show you. A demo video does exactly that, without the call. Recording them myself didn't scale. With agents writing the code, sometimes I can't keep up with my own PRs. On top of that I had to prepare the environment, record, edit, add zooms and do it all again after every fix. Now the agent does that part. The obvious answer was a hosted avatar product like HeyGen. It would look better than anything in this post. But I'm cheap, and I didn't want to pay for it. So I set four rules: - Free. No subscriptions, no credits. - Local. My face and voice don't go to anyone's servers. - Mac only. My work machine is a MacBook Pro with an M4 Max and 36 GB of memory. There's no NVIDIA GPU, so anything that needs CUDA was out. - The real app. No mock-ups and no clips cut from a marketing site. If the video shows a click, the app got clicked. I didn't write most of the code. I worked with a coding agent (GitHub Copilot) that had computer use on my Mac. It installed the models, ran the benchmarks, drove the app and showed me what came out. My job was to watch and say what was wrong. A lot of this post is that list. !Pipeline: a script goes to the voice clone and to the screen take; the voice drives the avatar and gives timing cues to the screen take; the screen take and avatar are composed into the final demo video. One script in, one video out. The voice's word timings tell the screen recording when to act. Part 1: The face Hosted tools hit a wall fast Hosted avatar products like HeyGen look great, and their free plans are small. When I checked, HeyGen's free plan gave you three one-minute videos a month with a watermark. They also need your face on their servers. That broke two of my rules on day one, so I looked for models that run on a laptop. Three lip-sync models, one clip, one scoreboard Lip-sync means you start from a real video of yourself and regenerate only the mouth to match new audio. Your face, your room and your head movement stay real. I picked three models that can run on Apple Silicon and gave each the same 17-second clip of me (cut from a tech talk I'd given) and the same voice track. I didn't want to judge by eye alone, so the agent scored them: - SyncNet confidence: a model trained to spot out-of-sync video rates how sure it is that lips and audio match. Higher is better. - Offset: how many frames the lips run ahead of or behind the voice. - Still looks like me: face-recognition similarity to the original. 1.0 means identical. | Model | SyncNet confidence | Offset | Still looks like me | Time for the 17 s clip | | --- | --- | --- | --- | --- | | No lip-sync (baseline) | 0.38 | 11 frames | 1.00 | none | | MuseTalk 1.5 | 2.66 | 1 frame | 0.92 | 96 s | | FaceFusion (EDTalk) | 4.65 | 3 frames (120 ms) | 0.95 | 92 s, CPU only | | LatentSync 1.5 | 6.55 | 0 | 0.97 | about 9 min | !The same clip side by side: no lip-sync, MuseTalk 1.5, FaceFusion EDTalk and LatentSync 1.5, with mouth close-ups and scores under each. Same clip, same voice, three models. (It's a GIF, so no sound.) The numbers matched what I saw. This chart made it obvious: !Four rows, one per model. Blue shows voice loudness, red shows how open the mouth is. With no lip-sync and with MuseTalk the red line is almost flat; FaceFusion twitches; LatentSync opens and closes along with the voice and closes on p, b and m sounds. Red is how far my mouth opens, blue is how loud the voice is. Dotted lines mark words starting with p, b or m, where lips have to close. LatentSync won. Its newer 1.6 version works at 512 px, but one attention step alone needed about 8.6 GB and my Mac started swapping, so I settled on 1.5 at 256 px with a few small patches to run on Apple's GPU. Recording the source take The model needs footage of you to put new lips on. The agent built me a small recording page with a hands-free teleprompter: a neutral pause, a greeting, a bit of explaining, 30 seconds of just listening, and a sign-off. Each part gives the avatar different raw material to draw from. !The recording kit page: three shirts, about three minutes each, with a setup checklist. I recorded two takes in two shirts. Then it turned out the microphone had recorded pure noise, because QuickTime had picked the wrong input. It didn't matter. The model replaces the mouth anyway, and for timing the agent read my lip movement straight from the video with face landmarks instead of from the audio. Getting the head to move with the voice Good lips weren't enough. My head would nod during a pause in the new voice and sit still in the middle of a sentence, which looks like a bad dub. The fix was to nudge the source footage's speed a little (between 0.85× and 1.17×) and line up pauses in the new voice with moments where I'd paused on camera. One number tracks it: the correlation between voice loudness and mouth activity. It went from −0.08 without the nudge to 0.40 with it. Feedback that shaped the look - Circle to webcam tile. A round bubble looked like a sticker. A rounded 16:9 tile showing head, shoulders and a bit of room looks like someone on a call. - Colour. My camera's white balance turned a black T-shirt green. The fix is baked into the render. - No nodding. Nodding while nobody is talking reads as agreeing with no one. The avatar only uses the calmest stretch of the take. - Smaller, twice. The tile went from 352 px to about 260 px wide on a 1920 px video. Part 2: The voice Voicebox is an open-source app that clones a voice from a short sample and runs it locally. My first clone came from a meeting recording and sounded flat, a bit robotic. My second came from four lively bits of a tech talk where I was presenting, and the difference was huge. That's the one I use. What I learned about getting a cloned voice to sound like me on a good day: - Style instructions don't work on a cloned voice. "Say this enthusiastically" did nothing. Livelier lines did: a couple of exclamation marks, shorter sentences. - Generate the whole script in one go. Line by line, the energy jumped around: an excited first line, a sleepy third one. - Generate several takes and measure them. The agent makes six takes and keeps the one that matches the script word for word (checked with a transcript), has the most movement in pitch and loudness, and drops in pitch on the last word so the ending sounds finished. - Read every transcript. One take said "and then do anything" instead of "undo anything". That one sentence got swapped in from another take. - Don't respell names. Spelling "Annotator" phonetically to help the model backfired: it came out sounding French. The plain spelling plus picking the best take worked better. Three bugs hiding in the pauses The screen needs room between sentences, so the voice gets short pauses. Those pauses caused three bugs in a row. Bug 1: digital silence. The first version inserted silence between sentences. My feedback: "the hard pauses are too snappy". The sound dropped from my voice to absolute zero, and real rooms are never that quiet. Bug 2: the stuck record. The next version filled the gaps by looping a slice of the quiet bit between words. "It sounds snappy, repeated, like a stuck disk." Short loops are easy to hear. Bug 3: clipped words. Sentences were cut at the word edges a speech-to-text model (Whisper) gave us. In flowing speech those edges drift by about 0.3 seconds, so word endings got chopped. !Two audio level plots around the end of a sentence. Before: the level falls off a cliff to digital silence. After: it settles to a steady low room tone, then rises into the next word. The version that works cuts at the quietest point measured in the audio itself, fills gaps with smooth, non-repeating noise at about −66 dB, and never loops. An automated check now flags any drop of more than 30 dB within 15 ms anywhere in the track. Part 3: The screen Why it has to be the real app An early version stitched together clips from the product website. I rejected it: it had to be a recording of the installed app. The website shows what the app should do. A recording shows that it does. How the computer drives itself - A separate screen. DeskPad creates a virtual 1920×1200 display. The recording happens there, so my real screens stay mine, and ffmpeg records only that display. - Buttons by name. A small Swift tool posts mouse and keyboard events. It finds controls through the macOS accessibility API by their names ("Pen", "Arrow", "Laser pointer") instead of pixel positions, so a moved button doesn't break the take. - A cursor that moves like a hand. Moves take longer for far or small targets (Fitts's law), overshoot slightly and follow a curve. - Timing from the voice. The voice comes with word timings, so the plan can say "draw the box when I say box": at 16.20 # a beat before the word "box" tool Pen path 0.8 -1528 512 -1450 506 -1360 505 ... mark box # log when it really happened Every action logs when it really happened. If any lands more than about half a second from its word, the agent records another take. The first version took about 15 takes. !Six frames from one take, each captioned with the words being spoken: a messy box around the download button, a curvy arrow and a "Start here!" label, a laser loop around the Applications icon, numbered steps, the arrow turning blue, and the page scrolled back to the top. Six moments from one take, with what the voice is saying at each. What broke | What I saw | Why | Fix | | --- | --- | --- | | Stray "4" and "5" counters later in the video | The toolbar hides itself after 3 seconds. A click meant for it hit the page while the counter tool was still on. | Check the button is still there right before clicking. Switch back to Select straight after the last counter. | | Colour keys did nothing, the arrow stayed blue | My keyboard was on Hebrew, and the app's shortcuts read the typed character. | The recorder won't start unless the layout is English. | | Every click missed | I'd rearranged my monitors overnight. | A setup script finds the virtual display and shifts every coordinate. | | The laser pointer looked robotic | Constant speed between points makes sharp corners. | About 120 points along an eased curve, with equal time per point. | | Things changed like magic | Fast moments use keyboard shortcuts, which you can't see. | Show the keys on screen as keycaps. | | Undo looked like a glitch | Undoing a move made the arrow snap back. | Undo a throwaway scribble instead. | | The label landed 2 seconds late | A trip to the toolbar costs about 2 seconds. | Use the keyboard shortcut. | !Four keycaps: control + 2 "close", ⌘ + Z "undo" (pressed), L "laser" and B "blue". Keycaps appear at the right edge when a shortcut is pressed, so nothing changes by magic. A few smaller touches came from watching drafts. A card with the URL slides up when the voice says it. Captions come from the voice's word timings. The video ends on the website's promo reel, synced through Chrome's DevTools Protocol, so the last shot lands on "Draw over any app." Part 4: The 23-minute problem LatentSync's quality came at a price. It runs 20 diffusion steps per frame, and one minute of video took 23 minutes to render. I said it plainly at the time: "Even 23 minutes for 1 minute is too much. Sometimes we render 10 minutes and I don't want to wait 5 hours." So I went back to MuseTalk. It had the weakest lips in the benchmark, but it makes each frame in one network pass instead of 20 steps. Profiling it showed most of the time wasn't in the model at all: turning its output back into pixels took about 70% of GPU time. Three changes: 1. Swap the full image decoder for TAESD, a tiny decoder built for the same kind of output. 2. Prepare the face once per take (landmarks, encoded frames, blend masks) and cache it. 3. Write frames straight into the video encoder in a single pass, with no files in between. !Bar chart of minutes of compute to render one minute of avatar video: LatentSync 1.5 23.5 minutes, MuseTalk 1.5 with the stock decoder 2.6 minutes, MuseTalk 1.5 with the tiny decoder and cache 1 minute. | Setup | Minutes to render 1 minute of video | | --- | --- | | LatentSync 1.5, 20 diffusion steps | 23.5 | | MuseTalk 1.5, stock decoder | 2.6 | | MuseTalk 1.5, tiny decoder and cache | 1.0 | A 59-second video now renders in 61 seconds. !Left: LatentSync, 23 minutes to render this minute. Right: fast MuseTalk, 61 seconds. Below each, the same video at the real tile size. Old renderer on the left, new on the right, the same ten seconds at full size and at tile size. (No sound in the GIF.) At full size, the fast version's lips are softer. At the size of the tile, nobody noticed. The slow model is still there behind a --quality best flag. In the benchmark I judged the models full screen. The product is a 260-pixel tile. If I'd judged at the size people actually see, I'd have picked the fast model a week earlier. Part 5: How I use it on PRs The pipeline exists for pull requests. These are the rules I settled on, most of them from me watching a video and saying "no, not like that": - Every PR gets a demo in its description, without anyone asking. Stacked PRs each get their own, showing only their own changes. - Open on the app. No title slide, no agenda and no "Hey". End on a plain "Thanks". - Show, then explain. The real app in the browser, the cursor working, my voice explaining as it goes, like showing a colleague. - One short zoom-out, only when it's needed. For things you can't see in the UI: where new data is saved, who can read it, a new API route or background job. Taken from the code, not from memory. - Before and Now side by side for anything visible. The agent runs the base branch from a second worktree on another port, with the same demo data, the same clicks and the same pace. - Plain, casual English. Every script goes through a no-AI-slop pass before the voice is generated. - Demo data only. A made-up demo user, never real client data, and nothing from the recording gets committed. - Short and light. About a minute per main change and two to four minutes in total, at −16 LUFS loudness and under 90 MB so GitHub plays it inline. The PR description stays short. My words to the agent were roughly "only the intent, what's necessary for a human to read, and the visuals": Two or three lines: what changed and why. Figma: <link> **Demo** <video> | | Before | After | | -------------- | --------- | --------- | | Template editor | <image> | <image> | | PDF export | <image> | <image> | **Notes for reviewers** - the one or two things worth a closer look It isn't free in tokens. A full demo cost the agent 28 to 99 million tokens (about 59 million on average) and 44 to 128 minutes. An updated screenshot costs about 2 million tokens and five minutes. So after a review fix, the default is a new screenshot in the Before | After table. The agent re-records one beat only when a fix changes what the video shows, and the whole video only when I ask. Recording, voice and rendering also load the CPU and GPU, so only one demo runs on my machine at a time. The avatar is off by default for PR demos. My voice stays on. A face on screen means something when you're showing stakeholders, not when a reviewer is checking a change. Leaving it off also saves the processing power when it isn't needed. The same skill makes complete demo videos too, and that's where the avatar comes in: sprint demos and public videos like the Annotator one above, where a human face is worth it. What I'd do differently - Judge at the size people watch. It would have saved the whole LatentSync detour. - Record the source take with a working mic. Clean audio from the same session would have made a better voice clone than an old meeting recording. - Assume the UI will move. Toolbars hide, keyboards switch layouts and monitors get rearranged. Check the state before every step instead of trusting coordinates. I still have an open question. The avatar loops back and forth over the same calm stretch of footage, which is fine for two minutes, but I don't have a good answer yet for natural head movement in a 10-minute video. If you've solved that on a laptop, I'd love to hear how. Try it Everything here is free and runs on a Mac with Apple Silicon: - Voice: Voicebox (Qwen3-TTS 1.7B). - Lip-sync: MuseTalk 1.5 with the TAESD decoder. LatentSync if you have the time. - Recording: DeskPad for a virtual display, ffmpeg to record it, and macOS accessibility APIs to drive the app. - The app in the demo: Annotator, a free screen-drawing app for Mac and Windows. Here's the full Annotator video from the top of the post: {% embed youtu.be %}

Harvesting SWE Benchmarks from Wild Vibe Coding Sessions Development teams need reproducible tasks from their own repositories and workflows to evaluate coding agents. Conversational developer-agent collaboration ("vibe coding") produces abundant coding sessions and repository checkpoints, but requirements evolve across turns, unrelated conversations co-occur with commits, and engineering efforts can span multiple sessions. We present SessionFarmer, an automated framework that turns these records into self-contained, executable software engineering tasks. It analyzes dialogue and commit history to filter unrelated sessions and group multi-turn, potentially multi-session interactions into work episodes centered on one engineering goal. For each episode, it reconstructs a hermetic execution container, synthesizes a task statement from the full context, and authors repository-grounded engineering-quality rubrics and unit tests. Closed-loop test refinement uses requirement-breaking mutants and a blind solver to reduce false rejections of valid alternatives and false acceptances of incomplete solutions. We construct CropChat, an open benchmark of 242 tasks across 61 public repositories and 14 programming languages, and CropEnterprise from proprietary enterprise logs. In a CropEnterprise audit, the original developers endorse the vast majority of reconstructed episode boundaries, task statements, and quality rubrics. Full-episode context covers substantially more user requirements than single-turn or single-session inputs. Test refinement cuts false-rejection rates and raises balanced accuracy against reference patch labels by 7-9 percentage points across solvers. Across 12 model-harness configurations on CropChat, the strongest setup achieves 39.3% Pass@1, while harness choice shifts Pass@1 by up to 8.4 percentage points for the same model. Comments: Code: bytedance/SessionFarmer ; Benchmark: huggingface.co

DeerFlow per-user memory: issue #1772 is open, but main already isolates memory --- title: "DeerFlow per-user memory: issue #1772 is open, but main already isolates memory" published: true canonical_url: memorysync.hashnode.dev description: "DeerFlow issue #1772 asks for per-user memory. It's still open, but current main stores memory per user. How it works and what to check." tags: ai, python, agents, langgraph --- Disclosure: I work on MemorySync (I'm Mohammed, the founder). I drafted this page with AI assistance and checked it against the source issue and the DeerFlow source on its main branch (commit d4ed400, Oct 28, 2026). I read the code; I didn't run DeerFlow for this page. The question bytedance/deer-flow#1772, "Support per-user memory isolation for multi-tenant deployment", reports that DeerFlow's memory.json is global and shared across all users, and asks for per-user storage such as memory/{user_id}.json. A maintainer replied that memory shared across users "can easily get mixed up" in a multi-user setup and assigned it. Status (checked Oct 28, 2026): the issue is still open, last updated Apr 3, 2026. But the code has moved on. If you're reading the issue and assuming every user still shares one file, check your version first. > Skip the memory plumbing around this fix. Once the fix below is in, the MemorySync agent memory guide shows how to add per-user memory with MemorySync. MemorySync doesn't fix this error itself. When you're ready, see MemorySync. Founding offer: use code FOUNDING30 for 30% off Starter for your first 3 months (Settings → Billing). What current main does From the memory module's AGENTS.md and the DeerMem config in the repo (main, Oct 28, 2026): - The default backend (manager_class: deermem) stores memory per user: {base_dir}/users/{user_id}/memory.json {base_dir}/users/{user_id}/agents/{agent_name}/facts/... - memory.backend_config.storage_path is a root directory. Empty means DeerFlow's base dir. A non-empty value becomes the root, with per-user memory still under {storage_path}/users/{uid}/memory.json. If it points at an existing file (for example an old memory.json path from before), config validation raises and tells you to point it at a directory. - The user comes from server-side identity, not from the client. In order: the LangGraph Agent Server's authenticated user, langgraph_auth_user_id, the gateway's auth-validated runtime.context["user_id"], the auth middleware's context variable, and finally "default". The Gateway strips client-supplied langgraph_auth_user and langgraph_auth_user_id before the graph is built. - Background memory updates capture user_id when the work is queued, so the identity survives the debounce timer. The catch: no-auth mode is still one shared user Per AGENTS.md, no-auth mode uses DEFAULT_USER_ID, which is "default". If you run DeerFlow as a service without authentication, every caller resolves to the same user and shares the same users/default/ memory, which is the behaviour the issue describes. Per-user isolation only helps once requests carry an authenticated identity. Checklist for a multi-user deployment 1. Turn on authentication so the gateway or LangGraph server supplies a real user. Without it, you're on "default". 2. Check storage_path. Leave it empty or point it at a directory. Don't point it at a .json file. 3. Migrate old global data. The repo ships backend/scripts/migrate_user_isolation.py. Run it with --dry-run first. It assigns legacy memory.json, agents/, skills/ and the global USER.md to --user-id (default default), and moves threads/ to their recorded owner. Pick the --user-id deliberately, because that user inherits everything that used to be shared. 4. Verify on disk. After two different signed-in users chat, you should see two directories under users/. If you see only users/default/, identity isn't reaching memory. 5. Mind multiple workers. A separate open RFC, #2283, describes a last-writer-wins race when several worker processes write the same memory.json. Per-user files narrow it to one user's concurrent turns, but don't remove it. Common questions Is DeerFlow memory shared between users? On current main, not if requests are authenticated: memory lives under users/{user_id}/. In no-auth mode everyone is "default", so it is shared. Why is issue #1772 still open? The issue thread doesn't say. As of Oct 28, 2026 it has no closing comment, and its last update was Apr 3, 2026. Treat the source on your installed version as the answer, not the issue state. How do I move existing global memory to a user? Use backend/scripts/migrate_user_isolation.py --dry-run, then run it without --dry-run and with the --user-id you want the legacy data assigned to. Skip the plumbing with MemorySync If you want per-user memory in a managed memory service instead of files on the gateway's disk, MemorySync scopes every user-scoped memory call by an end-user ID (X-End-User-ID / end_user_id), and its docs recommend a stable opaque ID, not an email address. There's no DeerFlow integration, so you'd call it from your own code. Read the MemorySync authentication and scoping guide first. For an overview of MemorySync, see the MemorySync homepage. From the guide: import os from memorysync import MemorySyncClient client = MemorySyncClient( api_key=os.environ["MEMORYSYNC_API_KEY"], base_url="https://api.memorysync.io", project_id=os.environ.get("MEMORYSYNC_PROJECT_ID") or None, end_user_id="user-123", ) result = client.query("What constraints and prior decisions matter for this task?", k=5) for memory in result.memories: print(memory.text) Snippet adapted from the MemorySync agent memory guide; unverified (not run) for this page. It keeps the guide's base_url line because MemorySyncClient requires it. Starter template: memorysyncio/memorysync-first-memory. Sources - bytedance/deer-flow#1772, "Support per-user memory isolation for multi-tenant deployment" (open; read 2026-10-28) - bytedance/deer-flow#2283, "RFC: Single-Writer Memory Queue" (open; read 2026-10-28) - DeerFlow main @ d4ed400 (2026-10-28): backend/packages/harness/deerflow/agents/memory/AGENTS.md, .../backends/deermem/deermem/config.py, .../runtime/user_context.py, backend/scripts/migrate_user_isolation.py, config.example.yaml - MemorySync docs: docs.memorysync.io/api/authentication Next step: read the MemorySync agent memory guide, then see MemorySync. Founding offer: use code FOUNDING30 for 30% off Starter for your first 3 months (Settings → Billing).

புதிய குரலுக்கு பழைய வீடியோவின் உதடு ஒத்துப்போகாது. LatentSync உதட்டு அசைவை ஒலிக்கு ஏற்ப மாற்றுகிறது.. உங்கள் டப்பிங் வீடியோவில் இதை முயற்சி செய்வீர்களா ? bytedance/LatentSync Instagram-லும் பின்தொடருங்கள்: instagram.com

Agent TARS lets AI see and control your desktop Agent TARS lets AI see and control your desktop ​ Agent TARS is an open source multimodal AI agent stack that sees your screen and interacts with your desktop. It ships two projects, Agent TARS and UI TARS desktop, for anyone exploring AI computer use. ​ bytedance/UI-TARS-desktop ​ #AI #Opensource #AgentTARS #MultimodalAI ​ Follow for daily open source drops 👇 [ar] Agent TARS يتيح للذكاء الاصطناعي رؤية والتحكم في سطح مكتبك يُمكّن Agent TARS الذكاء الاصطناعي من رؤية سطح مكتبك والتحكم به. Agent TARS عبارة عن حزمة برمجية مفتوحة المصدر ومتعددة الوسائط تعتمد على الذكاء الاصطناعي، حيث ترى شاشتك وتتفاعل مع سطح مكتبك. تتضمن الحزمة مشروعين: Agent TARS وUI TARS desktop، لكل من يرغب في استكشاف استخدام الذكاء الاصطناعي في الحواسيب. bytedance/UI-TARS-desktop #الذكاء_الاصطناعي #مفتوح_المصدر #AgentTARS #الذكاء_الاصطناعي_متعدد_الوسائط تابعنا للحصول على تحديثات يومية مفتوحة المصدر 👇 [bn] Agent TARS-এর মাধ্যমে AI এখন আপনার ডেস্কটপ দেখতে এবং নিয়ন্ত্রণ করতে পারবে এজেন্ট টার্স এআই-কে আপনার ডেস্কটপ দেখতে ও নিয়ন্ত্রণ করতে দেয় এজেন্ট টার্স একটি ওপেন সোর্স মাল্টিমোডাল এআই এজেন্ট স্ট্যাক যা আপনার স্ক্রিন দেখতে পায় এবং আপনার ডেস্কটপের সাথে ইন্টারঅ্যাক্ট করে। যারা এআই কম্পিউটার ব্যবহার নিয়ে কাজ করতে আগ্রহী, তাদের জন্য এটি এজেন্ট টার্স এবং ইউআই টার্স ডেস্কটপ—এই দুটি প্রজেক্ট সরবরাহ করে। bytedance/UI-TARS-desktop #এআই #ওপেনসোর্স #এজেন্টটার্স #মাল্টিমোডালএআই প্রতিদিন নতুন ওপেন সোর্স পেতে ফলো করুন 👇 [de-DE] Agent TARS lässt die KI deinen Desktop sehen und steuern Agent TARS ermöglicht es KI, Ihren Desktop zu sehen und zu steuern. Agent TARS ist ein Open-Source-Stack für multimodale KI-Agenten, der Ihren Bildschirm erkennt und mit Ihrem Desktop interagiert. Er umfasst zwei Projekte: Agent TARS und UI TARS Desktop. Diese eignen sich für alle, die sich mit der Nutzung von KI am Computer beschäftigen. bytedance/UI-TARS-desktop #KI #OpenSource #AgentTARS #MultimodaleKI Folgen Sie uns für tägliche Open-Source-Neuheiten 👇 [es-US] Agent TARS permite que la IA vea y controle tu escritorio Agent TARS permite que la IA vea y controle tu escritorio. Agent TARS es una plataforma de IA multimodal de código abierto que ve tu pantalla e interactúa con tu escritorio. Incluye dos proyectos: Agent TARS y UI TARS Desktop, para quienes exploran el uso de la IA en la computadora. bytedance/UI-TARS-desktop #IA #CódigoAbierto #AgentTARS #IAMultimodal Síguenos para recibir novedades diarias de código abierto 👇 [fr-FR] Agent TARS permet à l'IA de voir et de contrôler votre bureau Agent TARS permet à l'IA de voir et de contrôler votre bureau. Agent TARS est une suite logicielle d'IA multimodale open source qui analyse votre écran et interagit avec votre bureau. Elle comprend deux projets : Agent TARS et UI TARS Desktop, destinés à tous ceux qui explorent l'utilisation de l'IA sur ordinateur. bytedance/UI-TARS-desktop #IA #OpenSource #AgentTARS #IAMultimodale Suivez-nous pour découvrir chaque jour de nouveaux projets open source 👇 [hi] Agent TARS की मदद से AI आपके डेस्कटॉप को देख और नियंत्रित कर सकता है एजेंट TARS की मदद से AI आपके डेस्कटॉप को देख और नियंत्रित कर सकता है। एजेंट TARS एक ओपन सोर्स मल्टीमॉडल AI एजेंट स्टैक है जो आपकी स्क्रीन को देखता है और आपके डेस्कटॉप के साथ इंटरैक्ट करता है। यह AI कंप्यूटर के उपयोग को समझने के इच्छुक लोगों के लिए दो प्रोजेक्ट, एजेंट TARS और UI TARS डेस्कटॉप, उपलब्ध कराता है। bytedance/UI-TARS-desktop #AI #ओपनसोर्स #एजेंटTARS #मल्टीमॉडलAI रोजाना ओपन सोर्स अपडेट के लिए फॉलो करें 👇 [id] Agent TARS membuat AI bisa melihat dan mengendalikan desktop Anda Agent TARS memungkinkan AI untuk melihat dan mengontrol desktop Anda ​ Agent TARS adalah tumpukan agen AI multimodal sumber terbuka yang melihat layar Anda dan berinteraksi dengan desktop Anda. Ia menyediakan dua proyek, Agent TARS dan UI TARS desktop, untuk siapa pun yang ingin menjelajahi penggunaan komputer berbasis AI. ​ bytedance/UI-TARS-desktop ​ #AI #Opensource #AgentTARS #MultimodalAI ​ Ikuti untuk rilis sumber terbuka harian 👇 [it] Agent TARS permette all'IA di vedere e controllare il tuo desktop Agent TARS permette all'IA di vedere e controllare il tuo desktop Agent TARS è uno stack di agenti IA multimodali open source che vede il tuo schermo e interagisce con il tuo desktop. Include due progetti, Agent TARS e UI TARS desktop, per chiunque voglia esplorare l'utilizzo dell'IA nei computer. bytedance/UI-TARS-desktop #IA #Opensource #AgentTARS #IAMultimodale Seguici per aggiornamenti quotidiani sui progetti open source 👇 [iw] Agent TARS מאפשר ל-AI לראות ולשלוט בשולחן העבודה שלך Agent TARS מאפשר לבינה מלאכותית לראות ולשלוט בשולחן העבודה שלכם ​ Agent TARS הוא חבילת סוכני בינה מלאכותית רב-מודאלית בקוד פתוח שרואה את המסך שלכם ומקיים אינטראקציה עם שולחן העבודה שלכם. הוא מציע שני פרויקטים, Agent TARS ו-UI TARS לשולחן העבודה, לכל מי שחוקר את השימוש במחשבים מבוססי בינה מלאכותית. ​ bytedance/UI-TARS-desktop ​ #AI #OpenSource #AgentTARS #MultimodalAI ​ עקבו אחר פרסומים יומיים של קוד פתוח 👇 [ja] Agent TARS：PC画面を認識し操作するAIエージェント Agent TARSは、AIがデスクトップを認識し、制御することを可能にします。 Agent TARSは、画面を認識し、デスクトップと対話するオープンソースのマルチモーダルAIエージェントスタックです。AIを活用したコンピュータ利用を探求するすべての人向けに、Agent TARSとUI TARSデスクトップという2つのプロジェクトを提供しています。 bytedance/UI-TARS-desktop #AI #オープンソース #AgentTARS #マルチモーダルAI 毎日更新されるオープンソース情報をお見逃しなく👇 [ko] 내 데스크톱을 보고 제어하는 AI, Agent TARS Agent TARS는 AI가 사용자의 데스크톱을 인식하고 제어할 수 있도록 해주는 오픈 소스 멀티모달 AI 에이전트 스택입니다. Agent TARS는 사용자의 화면을 인식하고 데스크톱과 상호 작용하는 오픈 소스 멀티모달 AI 에이전트 스택입니다. AI 컴퓨터 활용에 관심 있는 모든 사용자를 위해 Agent TARS와 UI TARS 데스크톱 두 가지 프로젝트를 제공합니다. bytedance/UI-TARS-desktop #AI #오픈소스 #AgentTARS #멀티모달AI 매일 새로운 오픈 소스 소식을 받으려면 팔로우하세요 👇 [ml] Agent TARS ഉപയോഗിച്ച് AI-ക്ക് നിങ്ങളുടെ ഡെസ്ക്ടോപ്പ് കാണാനും നിയന്ത്രിക്കാനും സാധിക്കും ഏജന്റ് TARS നിങ്ങളുടെ ഡെസ്ക്ടോപ്പ് കാണാനും നിയന്ത്രിക്കാനും AI-യെ അനുവദിക്കുന്നു ​ ഏജന്റ് TARS എന്നത് നിങ്ങളുടെ സ്ക്രീൻ കാണുകയും നിങ്ങളുടെ ഡെസ്ക്ടോപ്പുമായി സംവദിക്കുകയും ചെയ്യുന്ന ഒരു ഓപ്പൺ സോഴ്‌സ് മൾട്ടിമോഡൽ AI ഏജന്റ് സ്റ്റാക്കാണ്. AI കമ്പ്യൂട്ടർ ഉപയോഗം പര്യവേക്ഷണം ചെയ്യുന്ന ഏതൊരാൾക്കും ഏജന്റ് TARS, UI TARS ഡെസ്ക്ടോപ്പ് എന്നീ രണ്ട് പ്രോജക്ടുകൾ ഇത് അയയ്ക്കുന്നു. ​ bytedance/UI-TARS-desktop ​ #AI #Opensource #AgentTARS #MultimodalAI ​ ദിവസേനയുള്ള ഓപ്പൺ സോഴ്‌സ് ഡ്രോപ്പുകൾക്കായി പിന്തുടരുക 👇 [nl-NL] Agent TARS laat AI je desktop zien en besturen Agent TARS laat AI je bureaublad zien en besturen Agent TARS is een open source multimodale AI-agentstack die je scherm ziet en met je bureaublad interacteert. Het bevat twee projecten, Agent TARS en UI TARS desktop, voor iedereen die AI-toepassingen op computers wil onderzoeken. bytedance/UI-TARS-desktop #AI #Opensource #AgentTARS #MultimodalAI Volg voor dagelijkse open source-updates 👇 [pa] Agent TARS AI ਨੂੰ ਤੁਹਾਡੇ ਡੈਸਕਟੌਪ ਨੂੰ ਦੇਖਣ ਅਤੇ ਕੰਟਰੋਲ ਕਰਨ ਦਿੰਦਾ ਹੈ ਏਜੰਟ TARS AI ਨੂੰ ਤੁਹਾਡੇ ਡੈਸਕਟਾਪ ਨੂੰ ਦੇਖਣ ਅਤੇ ਕੰਟਰੋਲ ਕਰਨ ਦਿੰਦਾ ਹੈ ਏਜੰਟ TARS ਇੱਕ ਓਪਨ ਸੋਰਸ ਮਲਟੀਮੋਡਲ AI ਏਜੰਟ ਸਟੈਕ ਹੈ ਜੋ ਤੁਹਾਡੀ ਸਕ੍ਰੀਨ ਨੂੰ ਦੇਖਦਾ ਹੈ ਅਤੇ ਤੁਹਾਡੇ ਡੈਸਕਟਾਪ ਨਾਲ ਇੰਟਰੈਕਟ ਕਰਦਾ ਹੈ। ਇਹ AI ਕੰਪਿਊਟਰ ਵਰਤੋਂ ਦੀ ਪੜਚੋਲ ਕਰਨ ਵਾਲੇ ਕਿਸੇ ਵੀ ਵਿਅਕਤੀ ਲਈ ਦੋ ਪ੍ਰੋਜੈਕਟ, ਏਜੰਟ TARS ਅਤੇ UI TARS ਡੈਸਕਟਾਪ ਭੇਜਦਾ ਹੈ। bytedance/UI-TARS-desktop ​ #AI #Opensource #AgentTARS #MultimodalAI ​ ਰੋਜ਼ਾਨਾ ਓਪਨ ਸੋਰਸ ਡ੍ਰੌਪਸ ਲਈ ਫਾਲੋ ਕਰੋ 👇 [pl] Agent TARS pozwala AI widzieć i obsługiwać Twój pulpit Agent TARS pozwala sztucznej inteligencji widzieć i kontrolować Twój pulpit Agent TARS to multimodalny pakiet agentów AI o otwartym kodzie źródłowym, który widzi Twój ekran i wchodzi w interakcję z Twoim pulpitem. Zawiera dwa projekty: Agent TARS i UI TARS desktop, przeznaczone dla każdego, kto interesuje się wykorzystaniem sztucznej inteligencji na komputerze. bytedance/UI-TARS-desktop #AI #Opensource #AgentTARS #MultimodalAI Obserwuj, aby otrzymywać codzienne informacje o nowych materiałach open source 👇 [pt-BR] Agent TARS permite que a IA veja e controle sua área de trabalho O Agent TARS permite que a IA veja e controle sua área de trabalho. O Agent TARS é um conjunto de agentes de IA multimodal de código aberto que vê sua tela e interage com sua área de trabalho. Ele inclui dois projetos, o Agent TARS e o UI TARS desktop, para qualquer pessoa que explore o uso de IA em computadores. bytedance/UI-TARS-desktop #IA #CódigoAberto #AgentTARS #IAMultimodal Siga para receber novidades diárias sobre código aberto 👇 [ru] Agent TARS позволяет ИИ видеть ваш рабочий стол и управлять им Agent TARS позволяет ИИ видеть и управлять вашим рабочим столом Agent TARS — это открытый многофункциональный стек агентов ИИ, который видит ваш экран и взаимодействует с вашим рабочим столом. Он включает в себя два проекта: Agent TARS и UI TARS desktop, для всех, кто изучает использование ИИ в компьютерах. bytedance/UI-TARS-desktop #AI #Opensource #AgentTARS #MultimodalAI Подписывайтесь на ежедневные обновления открытого исходного кода 👇 [ta] உங்கள் கணினியைப் பார்க்கவும் கட்டுப்படுத்தவும் உதவும் Agent TARS AI ஏஜென்ட் டார்ஸ் உங்கள் டெஸ்க்டாப்பைப் பார்க்கவும் கட்டுப்படுத்தவும் செயற்கை நுண்ணறிவை (AI) அனுமதிக்கிறது ஏஜென்ட் டார்ஸ் என்பது ஒரு திறந்த மூல பன்முக செயற்கை நுண்ணறிவு முகவர் தொகுப்பாகும். இது உங்கள் திரையைப் பார்த்து, உங்கள் டெஸ்க்டாப்புடன் ஊடாடுகிறது. செயற்கை நுண்ணறிவு கணினிப் பயன்பாட்டை ஆராயும் எவருக்காகவும், இது ஏஜென்ட் டார்ஸ் மற்றும் UI டார்ஸ் டெஸ்க்டாப் ஆகிய இரண்டு திட்டங்களை வழங்குகிறது. bytedance/UI-TARS-desktop #AI #Opensource #AgentTARS #MultimodalAI தினசரி திறந்த மூல வெளியீடுகளுக்குப் பின்தொடரவும் 👇 [te] Agent TARSతో మీ డెస్క్‌టాప్‌ను చూసి నియంత్రించగల AI ఏజెంట్ TARS, AI మీ డెస్క్‌టాప్‌ను చూడటానికి మరియు నియంత్రించడానికి అనుమతిస్తుంది ఏజెంట్ TARS అనేది ఒక ఓపెన్ సోర్స్ మల్టీమోడల్ AI ఏజెంట్ స్టాక్, ఇది మీ స్క్రీన్‌ను చూస్తుంది మరియు మీ డెస్క్‌టాప్‌తో ఇంటరాక్ట్ అవుతుంది. AI కంప్యూటర్ వినియోగాన్ని అన్వేషించే వారి కోసం ఇది ఏజెంట్ TARS మరియు UI TARS డెస్క్‌టాప్ అనే రెండు ప్రాజెక్ట్‌లను అందిస్తుంది. bytedance/UI-TARS-desktop #AI #Opensource #AgentTARS #MultimodalAI రోజువారీ ఓపెన్ సోర్స్ అప్‌డేట్‌ల కోసం ఫాలో అవ్వండి 👇 [uk] Agent TARS дозволяє ШІ бачити ваш робочий стіл та керувати ним Agent TARS дозволяє ШІ бачити та керувати вашим робочим столом ​ Agent TARS — це стек агентів ШІ з відкритим кодом, який бачить ваш екран та взаємодіє з ним. Він постачається з двома проектами: Agent TARS та UI TARS desktop, для тих, хто досліджує використання ШІ на комп’ютері. ​ bytandance/UI-TARS-desktop ​ #ШІ #ВідкритийКод #AgentTARS #МультимодальнийШІ ​ Підпишіться на щоденні оновлення з відкритим кодом 👇

DeerFlow 2.0: ByteDance's Open-Source Super Agent Harness, Setup and Walkthrough A complete walkthrough of bytedance/deer-flow, the open-source super agent harness that orchestrates sub-agents, memory, and sandboxes, powered by extensible skills. We cover what DeerFlow 2.0 is, the tools around it, how to install and configure models, and how to run it with Docker or local development. Chapters: 0:00 Introduction 0:48 What DeerFlow Is 2:08 The Ecosystem Around DeerFlow 3:46 Installing DeerFlow 5:50 Configuring Models 8:38 Running the Application 10:25 Who DeerFlow Is For 11:19 Summary Link: - bytedance/deer-flow: bytedance/deer-flow Narrated with an AI clone of the creator's own voice. Facts come from each project's or site's own pages. #OpenSource #GitHub #AI

⭐ BYTE-DANCE выложили DraMac, и это бесплатная замена Midjourney, которой не хватает многим — 17k+ звёзд за считанные дни! Генерит 2K-картинки на consumer GPU без магии и подписок. Репо: bytedance/DreamCraft — забирайте, пока не сделали платным 😼

This Free GitHub Repo Turns One Question Into a Finished Project It's called DeerFlow, and it comes from ByteDance. Most AI answers once and stops. DeerFlow keeps going, researching, writing code and creating files, even on tasks that take hours. Get it free here: bytedance/deer-flow Follow for more AI tips. #github #opensource #aiagents #bytedance #aitools #shorts [ar] هذا المستودع المجاني على GitHub يحول سؤالاً واحداً إلى مشروع متكامل يُسمى هذا البرنامج DeerFlow، وهو من تطوير شركة ByteDance. معظم برامج الذكاء الاصطناعي تستجيب مرة واحدة وتتوقف، بينما يواصل DeerFlow العمل، فيُجري الأبحاث، ويكتب التعليمات البرمجية، ويُنشئ الملفات، حتى في المهام التي تستغرق ساعات. حمّله مجانًا من هنا: bytedance/deer-flow تابعنا للمزيد من نصائح الذكاء الاصطناعي. #github #opensource #aiagents bytedance #aitools #shorts [bn] এই ফ্রি গিটহাব রিপো একটি প্রশ্নের মাধ্যমেই সম্পূর্ণ প্রজেক্ট তৈরি করে দেয় এর নাম ডিয়ারফ্লো (DeerFlow), এবং এটি বাইটড্যান্স (ByteDance) থেকে এসেছে। বেশিরভাগ এআই (AI) একবার উত্তর দিয়েই থেমে যায়। ডিয়ারফ্লো চলতেই থাকে, গবেষণা করে, কোড লেখে এবং ফাইল তৈরি করে, এমনকি ঘণ্টার পর ঘণ্টা সময় লাগে এমন কাজও। এটি এখানে বিনামূল্যে পান: bytedance/deer-flow আরও এআই টিপসের জন্য ফলো করুন। #github #opensource #aiagents #bytedance #aitools #shorts [de-DE] Dieses kostenlose GitHub-Repo macht aus einer einzigen Frage ein fertiges Projekt Es heißt DeerFlow und stammt von ByteDance. Die meisten KIs antworten nur einmal und hören dann auf. DeerFlow arbeitet kontinuierlich weiter, recherchiert, schreibt Code und erstellt Dateien – selbst bei Aufgaben, die Stunden dauern. Hier kostenlos herunterladen: bytedance/deer-flow Folgen Sie uns für weitere KI-Tipps. #github #opensource #aiagents #bytedance #aitools #shorts [es-US] Este repositorio gratuito de GitHub convierte una sola pregunta en un proyecto terminado Se llama DeerFlow y es de ByteDance. La mayoría de las IA responden una vez y se detienen. DeerFlow continúa trabajando, investigando, escribiendo código y creando archivos, incluso en tareas que llevan horas. Descárgalo gratis aquí: bytedance/deer-flow Síguenos para más consejos sobre IA. #github #opensource #aiagents #bytedance #aitools #shorts [fr-FR] Ce dépôt GitHub gratuit transforme une simple question en un projet complet Il s'agit de DeerFlow, développé par ByteDance. Contrairement à la plupart des IA qui répondent une seule fois, DeerFlow poursuit ses recherches, écrit du code et crée des fichiers, même pour des tâches qui durent des heures. Téléchargez-le gratuitement ici : bytedance/deer-flow Suivez-nous pour plus d'astuces sur l'IA. #github #opensource #aiagents #bytedance #aitools #shorts [hi] यह फ्री GitHub रिपो एक सवाल को पूरे प्रोजेक्ट में बदल देता है इसे डीयरफ्लो कहा जाता है, और यह बाइटडांस द्वारा विकसित किया गया है। अधिकांश एआई एक बार जवाब देकर रुक जाते हैं। डीयरफ्लो लगातार काम करता रहता है, शोध करता है, कोड लिखता है और फाइलें बनाता है, यहां तक ​​कि उन कार्यों पर भी जिनमें घंटों लगते हैं। इसे यहां से मुफ्त में प्राप्त करें: bytedance/deer-flow अधिक एआई टिप्स के लिए फॉलो करें। #github #opensource #aiagents #bytedance #aitools #shorts [id] Repo GitHub Gratis Ini Mengubah Satu Pertanyaan Menjadi Proyek Jadi Namanya DeerFlow, dan berasal dari ByteDance. Sebagian besar AI hanya menjawab sekali dan berhenti. DeerFlow terus berjalan, melakukan riset, menulis kode, dan membuat file, bahkan untuk tugas yang memakan waktu berjam-jam. Dapatkan secara gratis di sini: bytedance/deer-flow Ikuti untuk tips AI lainnya. #github #opensource #aiagents #bytedance #aitools #shorts [it] Questo repository GitHub gratuito trasforma una domanda in un progetto finito Si chiama DeerFlow ed è sviluppato da ByteDance. La maggior parte delle IA risponde una sola volta e si ferma. DeerFlow, invece, continua a lavorare, effettuando ricerche, scrivendo codice e creando file, anche per attività che richiedono ore. Scaricalo gratuitamente qui: bytedance/deer-flow Seguici per altri consigli sull'IA. #github #opensource #aiagents #bytedance #aitools #shorts [iw] המאגר החינמי הזה ב-GitHub הופך שאלה אחת לפרויקט מוכן זה נקרא DeerFlow, וזה מגיע מ-ByteDance. רוב הבינה המלאכותית עונה פעם אחת ועוצרת. DeerFlow ממשיכה, חוקרת, כותבת קוד ויוצרת קבצים, אפילו במשימות שלוקחות שעות. הורידו אותו בחינם כאן: bytedance/deer-flow עקבו אחר טיפים נוספים לבינה מלאכותית. #github #opensource #aiagents #bytedance #aitools #shorts [ja] たった一つの質問でプロジェクトを完成させるGitHubの無料AIツール これはByteDanceが開発したDeerFlowというAIです。ほとんどのAIは一度回答すると停止しますが、DeerFlowは調査、コード作成、ファイル作成などを継続し、何時間もかかるタスクにも対応します。 こちらから無料で入手できます：bytedance/deer-flow AIに関するその他のヒントについては、フォローしてください。 #github #opensource #aiagents #bytedance #aitools #shorts [ko] 질문 하나로 결과물을 뚝딱, 무료 GitHub 오픈소스 추천 DeerFlow라고 불리는 이 도구는 ByteDance에서 개발했습니다. 대부분의 AI는 한 번 답변하고 멈추지만, DeerFlow는 몇 시간씩 걸리는 작업이라도 계속해서 조사하고, 코드를 작성하고, 파일을 생성합니다. 여기에서 무료로 다운로드하세요: bytedance/deer-flow 더 많은 AI 팁을 보려면 팔로우하세요. #github #opensource #aiagents #bytedance #aitools #shorts [ml] ഒരൊറ്റ ചോദ്യത്തിൽ നിന്ന് പൂർണ്ണമായ പ്രോജക്റ്റ് നിർമ്മിക്കുന്ന സൗജന്യ GitHub റിപ്പോസിറ്ററി ഇതിനെ DeerFlow എന്ന് വിളിക്കുന്നു, ഇത് ByteDance-ൽ നിന്നാണ് വരുന്നത്. മിക്ക AI-കളും ഒരിക്കൽ ഉത്തരം നൽകുകയും നിർത്തുകയും ചെയ്യുന്നു. മണിക്കൂറുകൾ എടുക്കുന്ന ജോലികളിൽ പോലും DeerFlow ഗവേഷണം നടത്തുകയും കോഡ് എഴുതുകയും ഫയലുകൾ സൃഷ്ടിക്കുകയും ചെയ്യുന്നു. ഇവിടെ സൗജന്യമായി നേടൂ: bytedance/deer-flow കൂടുതൽ AI നുറുങ്ങുകൾക്കായി പിന്തുടരുക. #github #opensource #aiagents #bytedance #aitools #shorts [nl-NL] Deze gratis GitHub-repo verandert één vraag in een kant-en-klaar project Het heet DeerFlow en is ontwikkeld door ByteDance. De meeste AI-systemen geven maar één antwoord en stoppen dan. DeerFlow blijft doorgaan, onderzoek doen, code schrijven en bestanden aanmaken, zelfs bij taken die uren duren. Download het hier gratis: bytedance/deer-flow Volg ons voor meer AI-tips. #github #opensource #aiagents #bytedance #aitools #shorts [pa] ਇਹ ਮੁਫਤ GitHub ਰਿਪੋ ਇੱਕ ਸਵਾਲ ਨੂੰ ਪੂਰੇ ਪ੍ਰੋਜੈਕਟ ਵਿੱਚ ਬਦਲ ਦਿੰਦੀ ਹੈ ਇਸਨੂੰ ਡੀਅਰਫਲੋ ਕਿਹਾ ਜਾਂਦਾ ਹੈ, ਅਤੇ ਇਹ ਬਾਈਟਡਾਂਸ ਤੋਂ ਆਉਂਦਾ ਹੈ। ਜ਼ਿਆਦਾਤਰ ਏਆਈ ਇੱਕ ਵਾਰ ਜਵਾਬ ਦਿੰਦਾ ਹੈ ਅਤੇ ਰੁਕ ਜਾਂਦਾ ਹੈ। ਡੀਅਰਫਲੋ ਘੰਟਿਆਂਬੱਧੀ ਲੱਗਣ ਵਾਲੇ ਕੰਮਾਂ 'ਤੇ ਵੀ, ਖੋਜ, ਕੋਡ ਲਿਖਣਾ ਅਤੇ ਫਾਈਲਾਂ ਬਣਾਉਣਾ ਜਾਰੀ ਰੱਖਦਾ ਹੈ। ਇਸਨੂੰ ਇੱਥੇ ਮੁਫ਼ਤ ਪ੍ਰਾਪਤ ਕਰੋ: bytedance/deer-flow ਹੋਰ ਏਆਈ ਸੁਝਾਵਾਂ ਲਈ ਫਾਲੋ ਕਰੋ। #github #opensource #aiagents #bytedance #aitools #shorts [pl] To darmowe repozytorium na GitHubie zamienia jedno pytanie w gotowy projekt Nazywa się DeerFlow i pochodzi od ByteDance. Większość SI odpowiada raz i przestaje. DeerFlow działa dalej, wyszukuje informacje, pisze kod i tworzy pliki, nawet w przypadku zadań, które zajmują wiele godzin. Pobierz za darmo tutaj: bytedance/deer-flow Obserwuj, aby uzyskać więcej wskazówek dotyczących SI. #github #opensource #aiagents #bytedance #aitools #shorts [pt-BR] Este repositório gratuito do GitHub transforma uma pergunta em um projeto finalizado Chama-se DeerFlow e é da ByteDance. A maioria das IAs responde uma vez e para. O DeerFlow continua, pesquisando, escrevendo código e criando arquivos, mesmo em tarefas que levam horas. Baixe gratuitamente aqui: bytedance/deer-flow Siga para mais dicas de IA. #github #opensource #aiagents #bytedance #aitools #shorts [ru] Этот бесплатный репозиторий на GitHub превращает один вопрос в готовый проект Это называется DeerFlow, и он разработан компанией ByteDance. Большинство ИИ отвечают один раз и останавливаются. DeerFlow же продолжает работу, исследуя, записывая код и создавая файлы, даже в задачах, занимающих несколько часов. Скачать бесплатно можно здесь: bytedance/deer-flow Подписывайтесь, чтобы получать больше советов по ИИ. #github #opensource #aiagents #bytedance #aitools #shorts [ta] ஒரே கேள்வியை முழுமையான திட்டமாக மாற்றும் இந்த இலவச GitHub களஞ்சியம் இதற்கு டீர்ஃப்ளோ (DeerFlow) என்று பெயர், இது பைட் டான்ஸ் (ByteDance) நிறுவனத்திடமிருந்து வருகிறது. பெரும்பாலான செயற்கை நுண்ணறிவுகள் (AI) ஒருமுறை பதிலளித்துவிட்டு நின்றுவிடுகின்றன. ஆனால் டீர்ஃப்ளோவோ, பல மணிநேரம் எடுக்கும் பணிகளில்கூட, தொடர்ந்து ஆராய்ச்சி செய்து, நிரல்களை எழுதி, கோப்புகளை உருவாக்கிக் கொண்டே இருக்கிறது. இதை இங்கே இலவசமாகப் பெறுங்கள்: bytedance/deer-flow மேலும் செயற்கை நுண்ணறிவு குறிப்புகளுக்குப் பின்தொடரவும். #github #opensource #aiagents #bytedance #aitools #shorts [te] కేవలం ఒక్క ప్రశ్నతో పూర్తి ప్రాజెక్ట్‌ను సిద్ధం చేసే ఉచిత GitHub రిపోజిటరీ ఇది దీనిని డీర్‌ఫ్లో (DeerFlow) అని పిలుస్తారు, మరియు ఇది బైట్‌డాన్స్ (ByteDance) వారిది. చాలా AIలు ఒకసారి స్పందించి ఆగిపోతాయి. కానీ డీర్‌ఫ్లో, గంటల సమయం పట్టే పనులలో కూడా, పరిశోధన చేస్తూ, కోడ్ రాస్తూ, ఫైళ్లను సృష్టిస్తూ నిరంతరం పనిచేస్తూనే ఉంటుంది. దీనిని ఇక్కడ ఉచితంగా పొందండి: bytedance/deer-flow మరిన్ని AI చిట్కాల కోసం ఫాలో అవ్వండి. #github #opensource #aiagents #bytedance #aitools #shorts [uk] Цей безкоштовний репозиторій на GitHub перетворює одне запитання на готовий проєкт Він називається DeerFlow і походить від ByteDance. Більшість ШІ відповідають один раз і зупиняються. DeerFlow продовжує працювати, досліджуючи, пишучи код і створюючи файли, навіть для завдань, які займають години. Завантажте його безкоштовно тут: bytandance/deer-flow Підпишіться, щоб отримати більше порад щодо ШІ. #github #opensource #aiagents #bytandance #aitools #shorts