When LLM latency spikes, find out where.
Tell whether the slowdown is the model or something around it. DeepCube puts your Phoenix traces in the full request lifecycle.
See where a failure spread
When something breaks, you see it in context: which service is affected, what it depends on, where the problem propagates. That's spatial observability.
Is DeepCube an Arize Phoenix alternative?
For seeing LLM calls in the context of your whole application, yes. For LLM evaluation, alerting and paging, no. Keep Phoenix for those and run DeepCube next to it. Visualize takes 200 traces and logs per second for the whole grid, so check that against your volume. DeepCube is a Windows app today, with an alpha on macOS. It shows a quiet consumer or an unanswered call as a gap, but it does not yet flag a service that stops reporting.
Intro
Arize Phoenix is an AI observability and evaluation platform. 40+ auto-instrumentations (source: github.com/Arize-ai/openinference, verified 2026-11-07), evaluation tools, and a free, self-hostable codebase (source-available under the Elastic License 2.0; source: github.com/Arize-ai/phoenix/blob/main/LICENSE, verified 2026-11-14). We respect that depth. Arize is now part of Dynatrace: on 2026-10-01 Dynatrace announced that it "has completed its acquisition of Arize" and that "Arize will continue supporting both Phoenix, its open-source platform, and AX, its enterprise platform" (source: Dynatrace press release, ir.dynatrace.com, dated 2026-10-01, verified 2026-11-09). This page is an honest look at where DeepCube takes a different approach, where Phoenix excels, and how the two can work together.
DeepCube is the product; spatial observability is the category. We are not an APM company, we are a spatial observability company.
LLM performance in the context of your entire application stack
LLM Evaluation, Built Around the LLM Call
When your LLM latency spikes, where is the bottleneck?
- 40+ auto-instrumentations (source: github.com/Arize-ai/openinference, verified 2026-11-07) covering OpenAI, Anthropic, LangChain, LlamaIndex, and more. Phoenix delivers deep LLM-specific visibility.
- But LLM calls do not execute in a vacuum. They depend on API gateways, vector databases, embedding services, and the application underneath.
- When your RAG pipeline slows down, is it the model, the retrieval layer, the database feeding context, or a noisy neighbor? Phoenix shows the spans you send it, retrievers included. DeepCube keeps those spans in the same view as your application's metrics and logs, where a noisy neighbor shows up.
One spatial view. Full application context. Less to reconstruct.
See Your LLM Calls in Full System Context
Topology as your primary investigation surface.
- 3D spatial topology: Your LLM services, the APIs that call them, the vector databases they query, and the relationships between all of them.
- When something breaks, you see it in context: which service is affected, what it depends on, where the problem propagates.
- LLM calls next to application traces: see both in one spatial view, with application-level visibility.
- Phantoms: when a call goes out and nothing ever reports back from the other side, or a consumer goes quiet while messages keep arriving, DeepCube draws the missing side where it should be, so you see what is missing, not only what is there. It does not yet flag a service that stops reporting.
Zero vendor-specific agents. Full OTel ecosystem.
Architecture: How We Differ
Application-level correlation vs trace-focused AI observability.
DeepCube provides spatial observability across distributed tracing, metrics, logs, and LLM observability in a single spatial environment. When your LLM latency spikes, DeepCube shows you whether the bottleneck is the model provider, your API gateway, the vector database, or an upstream dependency.
Phoenix shows you the trace of your AI application. DeepCube shows you that trace in the context of the entire request lifecycle and the application dependencies around it.
| Aspect | DeepCube | Arize Phoenix |
|---|---|---|
| Scope | Spatial observability across application telemetry and LLM observability | AI observability and evaluation, focused on traces (source: arize.com/docs/phoenix, arize.com/docs/phoenix/tracing/concepts-tracing/otel-openinference/signals, verified 2026-11-06) |
| Instrumentation | Standard OpenTelemetry SDKs (all signals) | OpenTelemetry-based (LLM spans via OpenInference) (source, verified 2026-11-10) |
| LLM Auto-Instrumentation | OpenTelemetry GenAI semantic conventions | 40+ auto-instrumentations (OpenInference) (source: github.com/Arize-ai/openinference, verified 2026-11-07) |
| Visualization | 3D spatial topology | Trace tree views and per-project metrics dashboards; AX adds agent path and graph views (source: arize.com/docs/phoenix/tracing/llm-traces/metrics, arize.com/docs/ax/security-and-settings/pricing-and-usage, verified 2026-11-06) |
| Service Topology | Auto-discovered 3D service map | Trace-focused. Arize says teams can use Datadog for full-stack operational monitoring (source: arize.com/docs/phoenix/tracing/concepts-tracing/otel-openinference/signals, arize.com/compare/arize-vs-datadog, verified 2026-11-06) |
| Application Metrics | Application metrics via OTel correlated with traces | Metrics derived from traces: latency, cost, tokens, errors and annotation scores. AX adds custom metrics, dashboards and monitors on all tiers (source: arize.com/docs/phoenix/tracing/llm-traces/metrics, arize.com/docs/ax/security-and-settings/pricing-and-usage, verified 2026-11-06) |
| LLM Evaluation | Not a DeepCube feature | Evaluators for hallucination, toxicity, relevance, plus custom evals (source, verified 2026-11-07) |
| Cross-Signal Correlation | Traces + metrics + logs unified in spatial view | Traces are the focus (source: arize.com/docs/phoenix/tracing/concepts-tracing/otel-openinference/signals, verified 2026-11-06) |
AI on every plan. Context across the whole stack.
AI: Tessa vs Alyx, PXI and Signal
Both ship AI agents. On Phoenix and on AX Free and Pro, the difference is what they can see.
| Capability | Tessa (DeepCube) | Arize Phoenix |
|---|---|---|
| AI Diagnosis | In DeepCube, Tessa helps you read what the traces show, with spatial context | In-product AI diagnosis exists on both products and works from trace data. Alyx is AX's built-in AI engineering agent, covering trace investigation, aggregation, custom metrics, eval authoring, dataset curation and prompt iteration. PXI (Phoenix Intelligence) is the equivalent agent built into open-source Phoenix; its docs label it beta (verified 2026-11-09). Signal runs automatic failure-mode detection on a schedule, metered at 10 issues/mo Free and 25 Pro. Signal reads traces from the selected project. On Enterprise, custom managed agents can also query Dynatrace. Phoenix also serves a Remote MCP Server (beta) at /mcp that gives Claude Code, Cursor and VS Code what the Phoenix REST API can do (source: arize.com/docs/ax/alyx, arize.com/docs/phoenix/pxi and arize.com/docs/phoenix/integrations/mcp, verified 2026-11-06) |
| Codebase Access | In DeepCube, your local checkout: read, search, propose edits | No codebase access on Free or Pro, which "cannot attach a repo, open fix PRs, or choose a custom AI provider". Arize's own docs describe PXI (beta) as "a coding agent pointed at your observability data instead of a codebase". On Enterprise, you can "attach a GitHub repo so Signal can open pull requests for proposed fixes" (source: arize.com/docs/ax/observe/signal and arize.com/docs/phoenix/pxi, verified 2026-11-09) |
| What the AI can reach in your source | In DeepCube, Tessa reads your code where it already lives, on your machine. No vendor GitHub App, and no copy of your repo stored in our cloud. What Tessa reads is relayed to the model for your question and not kept on our servers. By default the model sees your workspace's top-level file names and any agent instructions file the repo ships. Other file contents go only when she opens them. (Tessa docs, data security) | On Enterprise, Signal attaches a GitHub repo with a personal access token, or a GitHub App where your account offers one: you create a GitHub skill ("name + personal access token") and "enter the repository as owner/repo" (source: arize.com/docs/ax/observe/signal and arize.com/docs/ax/agents/connect-your-harness, verified 2026-11-09) |
| Code Fixes | Available on every plan (on Start, within Tessa's free allowance of 1 request per 6 hours). In DeepCube, Tessa helps you follow the issue to the source file in your connected workspace and proposes the change. You own the outcome. | Alyx works across your traces, datasets, experiments and evaluators, and is available on all AX plans including Free (source, verified 2026-11-06). Signal's investigations include a proposed fix, sometimes a code change, on every plan within the 10 and 25 issue allowances; acting on it is what is gated. Repo attach with fix PRs, Custom Managed Agents and custom AI providers are Enterprise-only (source, pricing, verified 2026-11-09) |
| Root Cause Analysis | In DeepCube, Tessa works across LLM, app and dependencies | Trace-based: PXI (beta) walks failing traces, Signal reads project traces (PXI, Signal, verified 2026-11-10) |
| Accountability Model | In DeepCube, Tessa works through your traces, logs and code on her own, then brings you the answer. You own the outcome | Agent-assisted: Alyx in AX, PXI in Phoenix (beta), Signal on a schedule. On Enterprise, Signal opens fix PRs against an attached GitHub repo. (source, verified 2026-11-09) |
Arize ships real AI agents: Alyx in AX, PXI in Phoenix (beta), and Signal, which proposes fixes on every plan; on Enterprise, Signal opens fix PRs against an attached GitHub repo. Working in your code is table stakes now, and Tessa does it too: in DeepCube, she works in your connected codebase workspace and proposes the fix. You own the outcome. On Phoenix and on AX Free and Pro, the difference is what the AI can see. When your RAG pipeline degrades, in DeepCube, Tessa helps you read the traces around the latency spike, in one spatial view.
Phoenix and AX give you AI-focused visibility and agents that work from your traces. DeepCube puts those traces inside the whole application, and, in DeepCube, Tessa works across both.
LLM tracing and application traces in one view.
What you would pay, and for what
How each one bills.
- What DeepCube covers: LLM tracing, application traces and logs, and, in DeepCube, Tessa's AI diagnosis. Phoenix focuses on traces of AI applications.
- Predictable pricing: Node count sets the bill, with a volume discount on Visualize. No per-trace or per-evaluation charges.
- AI included: Tessa is part of every DeepCube plan at no extra charge. AI is included in the per-node price: there is no AI add-on required and no per-query AI charge. On paid plans the bundled Tessa allowance is a per-node budget; when it is spent, Tessa keeps working on a lighter model rather than stopping. Every plan can run, in DeepCube, Code Fix and root-cause analysis (on Start, within Tessa's free allowance of 1 request per 6 hours); higher plans put a more capable model behind them, and Fuse, available now through sales, adds voice and agents.
| Capability | DeepCube | Arize Phoenix |
|---|---|---|
| Bills on | Per application instance (a node) | Phoenix: free to use, self-hosted. AX: a monthly plan plus trace spans and ingestion volume (source: arize.com/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-phoenix-and-arize, arize.com/pricing, verified 2026-11-06) |
| List rate | Visualize: $20 per node, with a volume discount | Phoenix: free to use. AX Free: $0. AX Pro: $50 per month, then $.0008 per additional span and $3 per additional GB. AX Enterprise: custom (source: arize.com/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-phoenix-and-arize, arize.com/pricing, verified 2026-11-06) |
| In the price | Logs and Tessa | Phoenix: every Phoenix feature, none gated by AX. AX: Signal on every plan (Free 10 issues a month, Pro 25, Enterprise unlimited), unlimited evaluations and experiments (source: arize.com/pricing, verified 2026-11-10; arize.com/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-phoenix-and-arize, arize.com/docs/ax/security-and-settings/pricing-and-usage, verified 2026-11-06) |
| Not in the price | Infrastructure monitoring, alerting and paging. Optional: Tessa Personal seats (Tessa E5, E25 and E50), support hours. A $99.99 fee applies when you change plan. | Phoenix: your team provisions and scales the deployment. AX Free and Pro: Custom Managed Agents, repo access, Data Fabric and custom code evaluators are Enterprise (source: arize.com/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-phoenix-and-arize, arize.com/docs/ax/security-and-settings/pricing-and-usage, verified 2026-11-06) |
| Plan limits | Visualize: 200 traces and logs per second for the whole grid, 30-day retention, shared infrastructure. Start: 50 per second, 7-day retention. Telemetry above the limit can be dropped, and your collector will not show an error. DeepCube is a Windows app today, with an alpha on macOS. | AX Free includes 25k spans a month, 1 GB, 15-day retention and 10 Signal issues a month. AX Pro includes 50k spans a month, 10 GB, 30-day retention and 25 Signal issues a month (source: arize.com/docs/ax/security-and-settings/pricing-and-usage, verified 2026-11-06) |
| What it is for | Runs next to your current tool for tracing failures. No self-serve plan is meant to stand in for it today. Analyze and Fuse are available now through sales, with node minimums of 25 and 50. | Phoenix: agent development and evaluation platform. AX: AI engineering platform for improving AI agents and applications (source: arize.com/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-phoenix-and-arize, arize.com/docs/ax, verified 2026-11-06) |
| Your number | Count your application instances, then see deepcube.ai/pricing | See arize.com/pricing |
| LLM Observability | Included | Free and self-hostable (source-available under the Elastic License 2.0; source: github.com/Arize-ai/phoenix/blob/main/LICENSE, verified 2026-11-14) / AX Pro from $50/mo (source: arize.com/pricing, verified 2026-11-07) |
| Application Traces and Logs | Included | Focused on AI tracing. Arize says teams can use Datadog for full-stack operational monitoring (source: arize.com/compare/arize-vs-datadog, verified 2026-11-06) |
| AI Assistant | Included (Tessa) | Alyx optimizes prompts and investigates traces in-platform on all plans including Free, working across traces, datasets, experiments and evaluators. Signal proposes fixes on every plan; repo attach, fix PRs and Custom Managed Agents are Enterprise-only (source: arize.com/pricing, verified 2026-11-07) |
| 3D Spatial Topology | Included | Trace tree views and per-project metrics dashboards (trace tree, metrics, verified 2026-11-10) |
DeepCube pricing from deepcube.ai/pricing. Arize Phoenix is source-available under the Elastic License 2.0 (source: github.com/Arize-ai/phoenix/blob/main/LICENSE, verified 2026-11-14); Arize platform pricing varies by plan. Verify current pricing at arize.com/pricing. All prices USD.
Keep Phoenix for evals. Add DeepCube for application context.
Where Phoenix Excels
Honest comparison: if LLM quality loops are your primary job, Phoenix earns its place.
- A deep evaluation framework: hallucination, toxicity, relevance, and custom evals are Phoenix's core strength. A team whose main job is iterating on LLM output quality gets real depth here.
- 40+ OpenInference auto-instrumentations (source: github.com/Arize-ai/openinference, verified 2026-11-07) make it fast to capture rich LLM spans across OpenAI, Anthropic, LangChain, LlamaIndex, and more.
- Alyx (Arize AX), a built-in AI engineering agent (source: arize.com/docs/ax/alyx, verified 2026-11-06) that investigates traces, authors evals and custom metrics, curates datasets and runs prompt experiments without leaving the platform, on every plan including Free. Phoenix has its own equivalent, PXI, in beta.
If your primary need is tightening the LLM prompt quality loop, you may legitimately prefer Phoenix's depth there. That is not a contradiction with spatial observability: DeepCube and Phoenix can be complementary. Keep Phoenix for evals, add DeepCube for full application context.
Already Using Phoenix? Add Application Context.
OpenTelemetry makes it simple.
- Keep Phoenix for evals: If you love Phoenix's evaluation framework, keep using it. DeepCube adds application metrics and logs around your traces.
- No new agent: Phoenix is built on OpenTelemetry via OpenInference, so the same instrumentation can export OTLP to both. In DeepCube your spans arrive as traces, next to your application telemetry.
- Run both side by side: Compare the experience. When you see the value of correlating LLM performance with application health, you will understand why application-level context matters.
- Easy exit: If DeepCube is not right for you, change one endpoint and its API key header. Your instrumentation stays exactly the same.
Dual-Destination Collector Config
receivers:
otlp:
protocols:
grpc:
http:
exporters:
otlp/deepcube:
endpoint: "https://otlp.deepcube.ai"
headers:
API-Key: "YOUR-API-KEY"
otlp/phoenix:
endpoint: "localhost:4317"
tls:
insecure: true
service:
pipelines:
traces:
receivers: [otlp]
exporters: [otlp/deepcube, otlp/phoenix]
metrics:
receivers: [otlp]
exporters: [otlp/deepcube]
logs:
receivers: [otlp]
exporters: [otlp/deepcube]
Standard OTel Collector config. LLM traces go to both. Application metrics go to DeepCube.
Ready to See Your LLM Calls in Full Context?
Start free with DeepCube. Your OTel instrumentation stays as it is: send it through a Collector or any OTLP/gRPC exporter.
Compare DeepCube against other tools | LLM observability comparison