Both cover tracing, experiments on datasets, online scoring, human review and prompt versioning. Braintrust is a standalone hosted product organised around experiments, with SDKs in six languages; Weave is part of CoreWeave's model platform and fits teams that already work in Weights & Biases. Whether you want an eval product on its own or next to model training usually decides it.[1][6]
Braintrust is an independent, hosted evaluation platform built around experiments, while W&B Weave is the LLM and agent toolkit inside Weights & Biases, which CoreWeave acquired in May 2025 and has sold inside CoreWeave Forge since September 2026. Both trace agents, run offline and online evaluations and version prompts. Weave can block or modify inputs and outputs through scorers called from Python code; Braintrust documents no run-time blocking but ships a gateway in public preview.
| Decision factor | Braintrust | W&B Weave |
|---|---|---|
| Ownership | Independent; $80M Series B in February 2026 | CoreWeave since May 2025; sold inside CoreWeave Forge since 30 September 2026 |
| Centre of gravity | Experiments, with Loop, an in-product analysis agent | Traced functions and evaluations, next to W&B model training |
| SDKs | TypeScript, Python, Go, Java, Ruby, C# | Python and TypeScript; TypeScript lacks guardrails, class-based scorers and cost tracking |
| Self-hosting | Enterprise only, data plane only; Braintrust runs the control plane | Self-hosted on Kubernetes with a Weave licence and a W&B Platform install |
| Runtime guardrails | None documented; scoring and alerts run after logging | Partial: scorers called from Python code can block or modify inputs and outputs |
| Gateway | Hosted gateway in public preview, no SLA | None documented |
| Free tier | Starter: 1 GB processed data, 10K scores, 14-day retention | Forge Free: 1 GB of Weave ingestion a month, up to 5 model seats |
| First paid tier | Pro: $249 a month, 5 GB, unlimited users | Forge Pro: from $60 a month, 1.5 GB, teams under 50 employees |
| Capability | Braintrust | W&B Weave | Prefactor |
|---|---|---|---|
| Observe | |||
| LLM call tracing | ✓ | ✓ | ✓ |
| Multi-step agent tracing (tool calls, nested spans) | ✓ | ✓ | ✓ |
| Cost tracking | ✓ | ✓ | ✓ |
| OpenTelemetry support | ✓ | ✓ | ✓ |
| Evaluate | |||
| Offline evals (datasets, experiments) | ✓ | ✓ | ✓ |
| Online evals on production traffic | ✓ | ✓ | ✓ |
| LLM-as-a-judge scoring | ✓ | ✓ | ✓ |
| Code-based evaluators | ✓ | ✓ | ✓ |
| Human review and annotation | ✓ | ✓ | ✓ |
| Prompt versioning and playground | ✓ | ✓ | — |
| Act on findings | |||
| Runtime controls on live agent actions | — | Partial | ✓ |
| Gateway or proxy in the request path | ✓ | Not stated | — |
| Licence and deployment | |||
| Open source | Partial | Partial | Partial |
| Self-hosting | Partial | ✓ | Partial |
Braintrust sells one thing: a hosted platform for evaluating and observing AI applications. It is independent, raised an $80M Series B in February 2026 and ships quickly, though many of its newer features, including Patterns, Topics and a thread view, are still in public preview.
Weave comes with a wider platform attached. CoreWeave completed its acquisition of Weights & Biases in May 2025, and on 30 September 2026 launched CoreWeave Forge, which bundles W&B Models, Weave and new CoreWeave tools such as Agent Lens, Sandboxes and Inference. W&B hosted accounts now sign in through Forge, and pricing, documentation and login have moved to CoreWeave domains. CoreWeave now recommends Agent Lens, a separate product in public preview, for agent tracing, so expect Weave's packaging to keep changing. If your team already tracks training runs in W&B, having traces and evaluations in the same place is a real advantage. If it does not, the rest of Forge is weight you will not use.
Braintrust starts from the experiment. You bring a dataset, define a task and scorers, run variants and compare. Loop, its in-product agent, reads results, writes scorers and runs evals, and since September 2026 it can run on a schedule. Workflow evaluations through provider batch APIs, in public preview, are promoted as roughly half the cost of standard runs. In July 2026 Braintrust also published behaviour specs, an open standard for judging agent trajectories.
Weave starts from instrumented code. A decorator traces your functions, and its Evaluation API runs a model over a dataset with scorers, with results compared to catch regressions and shared as leaderboards. Production traces are scored by monitors, which Weave's own docs now describe as a previous approach in favour of Signals. Both products have annotation queues for human review and a playground for prompt work.
Braintrust is hosted. On Enterprise you can run its data plane in your own AWS, GCP or Azure account, but Braintrust always operates the control plane: the UI, logins and metadata. Weave can be fully self-hosted on Kubernetes with ClickHouse and S3-compatible storage, but only with a Weave-enabled licence and an existing W&B Platform install. Neither has a free self-hosted path.
Language coverage differs more than the feature lists suggest. Braintrust has SDKs for TypeScript, Python, Go, Java, Ruby and C#. Weave has Python and TypeScript, and its TypeScript SDK does not support guardrails, class-based scorers or cost tracking. A TypeScript-first team will get more from Braintrust today.
On run-time checks the order reverses. Weave's guardrails are scorers you call from Python application code that can block or modify inputs and outputs before they reach users. Braintrust documents no blocking or approvals on live agent actions; its online scoring, alerts and Patterns judge traces after logging. Neither documents approvals on an agent's tool calls.
Both meter data volume, but not in the same way. As of 20 October 2026, Braintrust Starter is free with 1 GB of processed data, 10K scores and 14-day retention, and Pro is $249 a month with 5 GB, 50K scores and 30-day retention; extra data on Pro is $3 per GB, and every plan includes unlimited users. Weave is priced through CoreWeave Forge: Forge Free includes up to 5 model seats and 1 GB of Weave ingestion a month, and Forge Pro starts at $60 a month with up to 10 model seats and 1.5 GB, for teams under 50 employees. Extra Weave ingestion is listed at $0.10 per MB.
The entry price is lower on Forge, but the listed overage rates differ sharply, and processed data and ingestion may not be measured the same way. Model your monthly trace volume against Braintrust's pricing and CoreWeave Forge pricing before you commit.
We sell the layer this section describes. Read it with that in mind.
Stated plainly, since Prefactor publishes this page: Prefactor is an agent evaluation platform that scores each step of a run and acts on what it finds through runtime policies on tool calls and data access, approval holds for risky actions and a kill switch for a run, an agent or a team. Braintrust evaluates after the fact, and Weave's guardrails check content in Python code without approvals on agent actions, so the hold-and-stop part is the gap here. Prefactor has no prompt playground or prompt registry and its platform is hosted, so if prompt iteration is your main need, Braintrust or Weave is the better pick. See the full comparison of 15 platforms.
Reviewed against public product and documentation pages on October 20, 2026. If a vendor has changed a feature, product name, or positioning since then, send a correction and we will update it. Numbered source links in the page body point to the ordered sources below.
Prefactor scores each step of every agent run and can hold or stop an action when a score says it should not happen. Start on the free tier: 5,000 spans a month, no credit card required.
Prefactor helps teams observe, evaluate, and improve their AI agents in production, across every framework and provider.