Blaxel vs CodeSandbox SDK
Both run each sandbox as a Firecracker microVM that sleeps with its memory and wakes with its processes still running. Blaxel wakes in under 25 ms and adds network rules on the provider side, while CodeSandbox SDK gives each VM a real disk and lets you fork a running sandbox that shares its memory.
The companies and products
Blaxel
Blaxel is a San Francisco company that runs cloud infrastructure for AI agents, and since September 2026 it is part of Baseten.
You get sandboxes, batch jobs, MCP server hosting and a model gateway. Each sandbox is a Firecracker microVM with its own kernel that drops to standby when nothing is connected and resumes in under 25 ms with its memory and running processes. It is built for agent products that start many sandboxes and want them asleep most of the time.
- Headquarters
- San Francisco, founded 2024
CodeSandbox SDK
CodeSandbox SDK is an API from CodeSandbox, now a Together AI company, that starts sandboxed development VMs on the same infrastructure as the CodeSandbox editor.
Each sandbox is a Firecracker microVM built from a template or any Dockerfile. When it goes idle it hibernates with a memory snapshot, and it resumes or forks from that snapshot in seconds. You drive it from a TypeScript SDK through terminals, commands, files and HTTPS hosts. Together's docs now call it the legacy offering and point new users to its successor, the together-sandbox SDK.
Which use cases each one fits
For each use case, Battleships prices every sandbox provider on the same workload and sorts them from the cheapest monthly bill up. “10th cheapest of 367” means 9 of the other 366 providers run that workload, with every feature it needs, for less. “With limits” means the provider runs it but misses something, so it is listed after every full match. Open a use case for its whole ranking and what each provider misses.
| Use case and its workload | Blaxel: price rank | CodeSandbox SDK: price rank | bochka: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Blaxel fits a software factory fully while CodeSandbox SDK is a compromise, and bochka adds SSH, /dev/kvm for Android emulators and a disk that stays. | 10th cheapest of 367 $2,373 a month | 178th of 367, with limits $1,939 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB | 12th cheapest of 366 $2,832 a month | 215th of 366, with limits $2,478 a month | 5th cheapest of 366 $657 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Both suit short code runs that sleep between calls, and Blaxel ranks higher here because it bills only while a connection is open and wakes in under 25 ms. | 14th cheapest of 345 $144 a month | 46th cheapest of 345 $642 a month | 186th of 345, with limits $2,000 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Neither fits 300 users at once on a self-serve plan: Blaxel stops at 200 and CodeSandbox SDK's Scale plan at 250, while bochka fits them on a self-serve plan. | 101st of 361, with limits $926 a month | 140th of 361, with limits $1,707 a month | Cheapest of 361 $209 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month | 77th of 360, with limits $4,378 a month | 348th of 360, with limits $3,861 a month | 34th of 360, with limits $2,000 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month CodeSandbox SDK fits agent evals fully while Blaxel's self-serve tiers cap how many run at once at 200, and bochka starts a fresh VM per task from a named snapshot. | 136th of 364, with limits $704 a month | 64th cheapest of 364 $1,335 a month | 10th cheapest of 364 $200 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month | 57th cheapest of 377 $431 a month | 92nd cheapest of 377 $860 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Neither has native SSH, so both are a compromise as a dev box you live in, while bochka is a plain machine you SSH into and leave on. | 265th of 368, with limits $114 a month | 310th of 368, with limits $240 a month | 18th cheapest of 368 $26 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Blaxel runs a desktop and a browser, CodeSandbox SDK has no desktop, and bochka streams a 60fps desktop with Chromium in every sandbox. | 14th cheapest of 369 $634 a month | 206th of 369, with limits $691 a month | 6th cheapest of 369 $200 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month Blaxel ships browser control, while on CodeSandbox SDK you build a headless Chromium template yourself; bochka has Chromium on a full desktop. | 16th cheapest of 367 $317 a month | 251st of 367, with limits $443 a month | 7th cheapest of 367 $200 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 16th cheapest of 346 $215 a month | 281st of 346, with limits $508 a month | 6th cheapest of 346 $90 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 147th of 338, with limits $4,292 a month | 193rd of 338, with limits $7,390 a month | 71st of 338, with limits $1,800 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month Blaxel fits well with domain-level egress rules on the provider side, and bochka gives root with raw TCP and nftables inside the VM. | 3rd cheapest of 360 $217 a month | 240th of 360, with limits $326 a month | 2nd cheapest of 360 $200 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month | Can't run it | Can't run it | Can't run it |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Can't run it | Can't run it | Can't run it |
Key features
| Feature | Blaxel | CodeSandbox SDK | bochka |
|---|---|---|---|
| Isolation | Firecracker microVM with its own kernel | Firecracker microVM with its own kernel | Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts |
| Sleep and wake | Standby a few seconds after the last connection; wakes in under 25 ms with memory and processes | Hibernates with a memory snapshot after an idle timeout of up to one day; resumes with processes running | Disk saved every minute and on stop; a stopped sandbox costs nothing and resume boots it again |
| Fork | Fork with memory and running processes (private preview) | Fork a hibernated sandbox from its snapshot, or a running one sharing its memory | Fork a running sandbox with its disk |
| Disk | Files live in memory (about half the RAM); one block volume can be attached at creation | 20 GB disk per VM; price for more disk not published | A real disk that stays, snapshotted every minute and on stop |
| Nested virtualization | No: Firecracker guests do not get /dev/kvm | Not documented | Yes: /dev/kvm, so Android emulators and VMs run inside |
| Shell access | Root, terminal through the bl connect command, no native SSH | Terminals, commands and files through the SDK, no SSH | Real SSH, root and systemd, plus exec and files over the API |
| Docker inside | Yes, with a Docker-in-sandbox image; custom images from Docker images | Yes, with root; sandboxes built from any Dockerfile | Yes, with systemd and root; templates from named snapshots, no Dockerfile images |
| Inbound traffic | Preview URL per port, public or token-protected; no raw TCP or public IP | HTTPS host per port, public by default, signed tokens for private hosts; no raw TCP or public IP | HTTPS URLs for up to 50 ports, token-protected by default; raw TCP; own IPv6 or IPv4 |
| Outbound rules | Domain allow and deny lists, method and path rules, static egress IPs | No documented egress allow lists or static egress IPs | Open internet; filter inside the VM with nftables |
| Keeping API keys out | Proxy fills secrets into outbound HTTPS requests (public preview) | Secrets passed in as environment variables | Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out |
| Desktop and browser | Desktop and browser control | No desktop; a headless Chromium template | Linux desktop streamed at 60fps, Chromium and computer-use tools |
| Coding agents | Claude Code image you drive from its terminal or your own harness | Run your own agent through the command API | Claude Code, Codex, OpenCode, pi and more, driven by a prompt, steer and events API |
| Sandboxes at once (self-serve) | 10, 50 or 200 by tier, standby sandboxes included | 10 on Build and Pro, 250 on Scale | 100 to 2,000 by plan, with a limit on starts per minute |
| Regions | US (Oregon, North Virginia), London, Frankfurt | A country hint steers scheduling, for example to Europe, but placement is not guaranteed | EU only: Germany, Finland, France, with EU data residency |
Every feature Battleships tracks (43)
| Feature | Blaxel | CodeSandbox SDK | bochka |
|---|---|---|---|
| Full VM (own kernel) | Yes | Yes | Yes |
| Docker inside | Yes | Yes | Yes |
| Nested virtualization | No | No | Yes |
| Root | Yes | Yes | Yes |
| Systemd | No | Not published | Yes |
| Desktop GUI | Yes | No | Yes |
| Browser | Yes | No | Yes |
| Computer-use API | Yes | No | Yes |
| SSH | No | No | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | Yes | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Yes | Yes | Yes |
| Memory fork | Yes | Yes | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | Yes | No |
| Wake on request | Yes | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | No | No | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | Yes | No | No |
| Secret proxy | Yes | No | Yes |
| Secret proxy for any API | Yes | No | Yes |
| Preinstalled agents | Yes | No | Yes |
| Agent harness API | No | No | Yes |
| MCP server | Yes | No | No |
| GPU | No | No | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | No | No | No |
| BYOC | Yes | No | No |
| Open source | No | No | No |
| Scoped API keys | Yes | Not published | Yes |
| Spend limits | No | Not published | Yes |
| Usage per sandbox (API) | Yes | No | Yes |
| Audit logs | Not published | Not published | No |
| SSO | Yes | Yes | No |
| SOC 2 | Yes | Yes | No |
| HIPAA | Yes | No | No |
| EU data residency | Yes | No | Yes |
| Zero data retention | Yes | Not published | Yes |
Performance
On the StarSling hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU and 8 GB, run on September 29, 2026, bochka did 26.69 runs per second and Blaxel 19.48, so bochka got about 1.4 times the work done in the same time. CodeSandbox SDK was not part of that run, so there is no comparable CPU score for it. Higher is better.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Blaxel measured on: Sandbox (active time, RAM-only pricing, CPU bundled). CodeSandbox SDK has no published run in this suite yet, so there is no like-for-like number to compare.
Pricing in five scenarios
Each bill is priced from the providers' public price lists by the open Battleships engine, with the cheapest plan that runs the scenario.
| Scenario | Blaxel, a month | CodeSandbox SDK, a month | bochka, a month |
|---|---|---|---|
| One coding agent, workdays 4 vCPU / 8 GB, 50 GB disk, 8 hours a day, 22 days a month, files kept between days | $33 · Pay-as-you-go Tier 1 | $193 · Scale | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $120 · Pay-as-you-go Tier 1 | $341 · Scale | $26 · $20/mo |
| 50 agents in parallel 50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside | $1,489 · Pay-as-you-go Tier 1 | $2,500 · Scale | $317 · $20/mo × 3 seats |
| Sandboxes for your users 2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once | $396 · Pay-as-you-go Tier 2 · with limits | $815 · Scale · with limits | $200 · $100/mo × 2 seats |
| 1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month | $876 · Pay-as-you-go Tier 2 · with limits | $1,632 · Scale · with limits | $2,000 · $2000/mo · with limits |
Blaxel: Blaxel charges for allocated RAM per second while a sandbox is active, with CPU included (8 GB comes with 4 cores). In standby you pay only snapshot storage, and self-serve tiers are unlocked by topping up your wallet with $20 or $50 of usage credit.
CodeSandbox SDK: CodeSandbox SDK sells VM credits that each size burns per hour, billed per started minute while the VM runs, including idle time before it hibernates. The Scale plan adds a $170 monthly platform fee that is not usage credit, and a hibernated VM costs nothing.
CodeSandbox SDK: Plans start at $20 a month, and the whole fee is usage credit billed per second while a sandbox runs. A stopped sandbox costs nothing and keeps its disk.
Limits in those scenarios
Blaxel's self-serve tiers allow 10, 50 or 200 sandboxes at once, counting those in standby, and CodeSandbox SDK's Scale plan allows 250, so neither fits 300 users or 1,000 short jobs at once without a sales plan. Blaxel's lower tiers also delete sandboxes after 7 or 30 days. CodeSandbox SDK gives each VM 20 GB of disk and publishes no price for more disk or for egress. bochka fits 300 users on a self-serve plan, but it limits starts per minute, so 1,000 jobs at the same moment get queued.
| Scenario | Blaxel | CodeSandbox SDK | bochka |
|---|---|---|---|
| One coding agent, workdays | None that bite |
| None that bite |
| One agent that never stops | None that bite |
| None that bite |
| 50 agents in parallel | None that bite |
| None that bite |
| Sandboxes for your users |
|
| None that bite |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use Blaxel when
your sandboxes sit idle most of the time and must wake in milliseconds, when you need domain-level egress rules or static egress IPs, or when they must run in the US.
- Wakes in milliseconds. For sandboxes that sit idle most of the time: one sleeps seconds after the last connection and wakes in under 25 ms.
- Network rules on the provider side. For teams that must fence agents in: allow or deny domains, down to methods and paths, with static egress IPs.
- US and UK regions. For products with users in the US: Blaxel runs in Oregon and North Virginia as well as London and Frankfurt.
Use CodeSandbox SDK when
you want to fork many runs from one warm, running state or need a real disk per VM, and you are fine building on what Together now calls its legacy SDK.
- Forks that share memory. For branching many runs from one warm state: fork a running sandbox while it shares memory with its parent.
- A disk in every VM. For repos and dependencies that outgrow memory: each VM gets a 20 GB disk instead of files held in RAM.
- Big machines on Enterprise. For very large single jobs: Scale goes to 16 cores and 32 GB, and Enterprise to 64 cores and 128 GB.
Where bochka fits
Blaxel and CodeSandbox SDK are both built around sandboxes that sleep with their memory and wake fast. If your agents instead work for hours on builds, tests, Docker and emulators, look at bochka: a whole Linux machine with a real disk, SSH, a desktop and the agents already installed.
A whole Linux machine
Blaxel guests get no /dev/kvm and CodeSandbox SDK does not document it, so Android emulators and VMs do not start. Neither has SSH, so editors and scp go through bl connect or SDK calls instead.
A disk that stays
Blaxel keeps files in memory, about half the RAM, unless you attach a volume at creation. CodeSandbox SDK gives 20 GB with no published price for more. bochka saves a real disk every minute, and a stop is free.
Agents, desktop and keys built in
On both you build the agent harness yourself, and CodeSandbox SDK has no desktop and puts keys in environment variables. bochka ships Claude Code, Codex and more with a prompt and steer API, a 60fps desktop and a credentials proxy.
Use bochka when your agents do real work for hours: builds, tests, Docker and emulators on a whole Linux machine with SSH, a disk that stays, a faster CPU and the agents and their keys already set up. Compare it one to one: bochka vs Blaxel and bochka vs CodeSandbox SDK.
Questions people ask
Is Blaxel cheaper than CodeSandbox SDK?
In all five scenarios we priced, yes. One coding agent on workdays costs $33 a month on Blaxel and $193 on CodeSandbox SDK, and 50 agents in parallel cost $1,489 against $2,500. bochka comes to $20 and $317 for the same two.
Which is faster, Blaxel or CodeSandbox SDK?
For CPU work there is no comparable score for CodeSandbox SDK. Blaxel did 19.48 runs per second on the Node.js benchmark and bochka 26.69. For waking up, Blaxel resumes from standby in under 25 ms.
Can Blaxel and CodeSandbox SDK run Docker?
Yes. Blaxel runs Docker with its Docker-in-sandbox image, and CodeSandbox SDK builds sandboxes from any Dockerfile and runs Docker inside. An Android emulator needs /dev/kvm, which Blaxel guests do not get and CodeSandbox SDK does not document, while bochka exposes it in every sandbox.
Can I SSH into Blaxel or CodeSandbox SDK?
Neither has native SSH. Blaxel gives a terminal through the bl connect command, and CodeSandbox SDK gives terminals, commands and files through its SDK. Every bochka sandbox has real SSH as well as an exec and files API.
Is CodeSandbox SDK still maintained?
CodeSandbox is now part of Together AI. Together's docs call the CodeSandbox-based SDK deprecated and name together-sandbox as its successor, while self-serve sign-up still runs through CodeSandbox.
What happened to Blaxel after Baseten?
Blaxel joined Baseten on September 10, 2026. Baseten says the product and the support team stay the same and features keep shipping.
Try bochka
Install the CLI, then run bochka new.
Product preview