feat(fireworks): add Fireworks AI provider by SantiagoDePolonia · Pull Request #475 · ENTERPILOT/GoModel

SantiagoDePolonia · 2026-07-04T16:28:47Z

Summary

Adds Fireworks AI (api.fireworks.ai/inference/v1) as a first-class provider. Setting FIREWORKS_API_KEY is the whole setup — the provider is auto-discovered, with optional FIREWORKS_BASE_URL and FIREWORKS_MODELS overrides following the standard convention.

User-visible impact

New provider type fireworks: chat completions, streaming, /v1/models discovery, embeddings, passthrough, and /v1/responses (translated via chat completions).
Model IDs are Fireworks' account-scoped paths (e.g. accounts/fireworks/models/gpt-oss-120b) and pass through verbatim, including in FIREWORKS_MODELS lists.
Model metadata, pricing enrichment, and cost tracking work out of the box: the fireworks type matches the ai-model-list registry key (76 models), and usage costing falls back to the shared OpenAI-compatible token mappings.

Provider-specific behavior

Implemented as a minimal openai.ChatCompatible embed (same shape as Z.ai/MiniMax/Xiaomi) — all transport is shared code; there is no Fireworks-specific request/response translation.
No native files/batches/audio surface is exposed (Fireworks' inference API doesn't offer OpenAI-compatible endpoints for them); tests pin this interface surface.
No dedicated docs page, per the overview's "set an API key and go" convention — documented via the provider matrix row + a note about account-scoped model IDs.

Testing

Unit tests: bearer auth + endpoint paths for chat/embeddings/models, interface-surface guards.
Full go test ./..., make lint (0 issues), and pre-commit make test-race all green.
E2E against a mock upstream: booted the gateway with only env vars — provider registered, /v1/models listed fireworks/accounts/fireworks/models/..., and chat/embeddings/responses round-tripped correctly with slash-containing model IDs.
Probed the live GET /inference/v1/models endpoint (401 without key) to confirm upstream discovery support.

🤖 Generated with Claude Code

Add Fireworks AI (api.fireworks.ai/inference/v1) as a provider, discovered via FIREWORKS_API_KEY with optional FIREWORKS_BASE_URL/FIREWORKS_MODELS. The provider embeds openai.ChatCompatible: chat completions, streaming, /models discovery, embeddings, and passthrough come from the shared OpenAI-compatible transport, with /v1/responses translated through chat. The "fireworks" type matches the ai-model-list registry key, so model metadata, pricing enrichment, and cost tracking work without extra mapping; account-scoped model IDs (accounts/fireworks/models/...) pass through verbatim. Document the provider in the overview matrix (no dedicated page needed: bearer key + optional base URL), .env.template, config.example.yaml, and the swagger description. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

mintlify · 2026-07-04T16:28:50Z

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated (UTC)
gomodel	🟢 Ready	View Preview	Jul 4, 2026, 4:29 PM

💡 Tip: Enable Workflows to automatically generate PRs for you.

coderabbitai · 2026-07-04T16:29:06Z

Warning

Review limit reached

@SantiagoDePolonia, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 16 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro

Run ID: b85a46ae-2632-4373-988a-9626b65929c7

📥 Commits

Reviewing files that changed from the base of the PR and between 2417ed1 and 20ada98.

📒 Files selected for processing (10)

.env.template
CLAUDE.md
README.md
cmd/gomodel/docs/docs.go
cmd/gomodel/main.go
config/config.example.yaml
docs/openapi.json
docs/providers/overview.mdx
internal/providers/fireworks/fireworks.go
internal/providers/fireworks/fireworks_test.go

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch feat/fireworks-provider

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

codecov-commenter · 2026-07-04T16:31:37Z

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

❌ Patch coverage is 45.45455% with 6 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
internal/providers/fireworks/fireworks.go	50.00%	5 Missing ⚠️
cmd/gomodel/main.go	0.00%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

greptile-apps · 2026-07-04T16:42:09Z

Confidence Score: 4/5

Safe to merge — the change is purely additive, isolated in its own package, and follows the established provider pattern exactly.

The implementation correctly embeds ChatCompatible and inherits all transport behavior. The only gap is a missing StreamChatCompletion test; every other observable behavior is covered. Doc updates are accurate: /responses works via the inherited chat-translation path, and passthrough is available when explicitly enabled.

internal/providers/fireworks/fireworks_test.go — could benefit from a streaming smoke test.

Sequence Diagram

%%{init: {'theme': 'neutral'}}%%
sequenceDiagram
    participant Client
    participant Gateway as GoModel Gateway
    participant Fireworks as Fireworks AI<br/>(api.fireworks.ai/inference/v1)

    Client->>Gateway: "POST /v1/chat/completions<br/>model: accounts/fireworks/models/gpt-oss-120b"
    Gateway->>Gateway: Route to fireworks.Provider
    Gateway->>Fireworks: "POST /chat/completions<br/>Authorization: Bearer fw_..."
    Fireworks-->>Gateway: ChatResponse
    Gateway-->>Client: OpenAI-compatible response

    Client->>Gateway: POST /v1/responses
    Gateway->>Gateway: providers.ResponsesViaChat()
    Gateway->>Fireworks: POST /chat/completions (translated)
    Fireworks-->>Gateway: ChatResponse
    Gateway-->>Client: ResponsesResponse (normalized)

    Client->>Gateway: POST /v1/embeddings
    Gateway->>Fireworks: "POST /embeddings<br/>Authorization: Bearer fw_..."
    Fireworks-->>Gateway: EmbeddingResponse
    Gateway-->>Client: OpenAI-compatible embedding response

    Client->>Gateway: GET /v1/models
    Gateway->>Fireworks: GET /models
    Fireworks-->>Gateway: Model list (account-scoped IDs)
    Gateway-->>Client: Normalized models list

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
sequenceDiagram
    participant Client
    participant Gateway as GoModel Gateway
    participant Fireworks as Fireworks AI<br/>(api.fireworks.ai/inference/v1)

    Client->>Gateway: "POST /v1/chat/completions<br/>model: accounts/fireworks/models/gpt-oss-120b"
    Gateway->>Gateway: Route to fireworks.Provider
    Gateway->>Fireworks: "POST /chat/completions<br/>Authorization: Bearer fw_..."
    Fireworks-->>Gateway: ChatResponse
    Gateway-->>Client: OpenAI-compatible response

    Client->>Gateway: POST /v1/responses
    Gateway->>Gateway: providers.ResponsesViaChat()
    Gateway->>Fireworks: POST /chat/completions (translated)
    Fireworks-->>Gateway: ChatResponse
    Gateway-->>Client: ResponsesResponse (normalized)

    Client->>Gateway: POST /v1/embeddings
    Gateway->>Fireworks: "POST /embeddings<br/>Authorization: Bearer fw_..."
    Fireworks-->>Gateway: EmbeddingResponse
    Gateway-->>Client: OpenAI-compatible embedding response

    Client->>Gateway: GET /v1/models
    Gateway->>Fireworks: GET /models
    Fireworks-->>Gateway: Model list (account-scoped IDs)
    Gateway-->>Client: Normalized models list

_{Reviews (1): Last reviewed commit: "Merge branch 'main' into feat/fireworks-..." | Re-trigger Greptile}

greptile-apps · 2026-07-04T16:42:13Z

+func TestProvider_DoesNotExposeOptionalOpenAICompatibleInterfaces(t *testing.T) {
+	provider := NewWithHTTPClient("fw-key", "", nil, llmclient.Hooks{})
+
+	if _, ok := any(provider).(core.NativeBatchProvider); ok {
+		t.Fatal("fireworks provider should not implement native batch provider")
+	}
+	if _, ok := any(provider).(core.NativeFileProvider); ok {
+		t.Fatal("fireworks provider should not implement native file provider")
+	}
+	if _, ok := any(provider).(core.AudioProvider); ok {
+		t.Fatal("fireworks provider should not implement audio provider")
+	}
+}


Missing streaming coverage

StreamChatCompletion is not exercised anywhere in the test suite. The chat test verifies the non-streaming path (/chat/completions), but the streaming path (/chat/completions with stream: true and an SSE response) is left untested. Other providers like MiniMax override streaming explicitly; fireworks inherits it but the test surface gap means a regression in ChatCompatible.StreamChatCompletion (or providers.EnsureChatCompletionSSE) would not be caught by this package's tests. A short TestStreamChatCompletion_UsesBearerAuthAndStreamEndpoint similar to the other two transport tests would close this gap.

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!

greptile-apps · 2026-07-04T16:49:23Z

Confidence Score: 4/5

Mostly safe to merge after fixing the passthrough allowlist mismatch.

Core chat, embeddings, model discovery, and registration follow the existing ChatCompatible provider pattern. One contained bug remains: Fireworks passthrough is documented and implemented on the provider, but disabled by the default server allowlist.

internal/providers/fireworks/fireworks.go, config/config.go, internal/server/passthrough_support.go

T-Rex Logs

What T-Rex did

The verifier confirmed that passthrough remains disabled, with only static code context inspected and no runtime artifacts captured because the step limit was reached.
A unit-test run for the fireworks provider produced a log showing the exact command and working directory used, but the run timed out before any test output or exit code could be captured, making validation inconclusive.

_{Ran code and verified through T-Rex}

Sequence Diagram

%%{init: {'theme': 'neutral'}}%%
sequenceDiagram
participant User
participant GoModel as GoModel Gateway
participant Factory as Provider Factory
participant Fireworks as Fireworks Provider
participant Upstream as api.fireworks.ai/inference/v1

GoModel->>Factory: Register fireworks.Registration
User->>GoModel: /v1/chat/completions or /v1/embeddings
GoModel->>Fireworks: Route model selector to provider
Fireworks->>Upstream: OpenAI-compatible request with Bearer FIREWORKS_API_KEY
Upstream-->>Fireworks: OpenAI-compatible response
Fireworks-->>GoModel: Normalized core response
GoModel-->>User: OpenAI-compatible response

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
sequenceDiagram
participant User
participant GoModel as GoModel Gateway
participant Factory as Provider Factory
participant Fireworks as Fireworks Provider
participant Upstream as api.fireworks.ai/inference/v1

GoModel->>Factory: Register fireworks.Registration
User->>GoModel: /v1/chat/completions or /v1/embeddings
GoModel->>Fireworks: Route model selector to provider
Fireworks->>Upstream: OpenAI-compatible request with Bearer FIREWORKS_API_KEY
Upstream-->>Fireworks: OpenAI-compatible response
Fireworks-->>GoModel: Normalized core response
GoModel-->>User: OpenAI-compatible response

_{Reviews (2): Last reviewed commit: "Merge branch 'main' into feat/fireworks-..." | Re-trigger Greptile}

greptile-apps · 2026-07-04T16:49:26Z

+// "accounts/fireworks/models/llama-v3p1-8b-instruct" and pass through
+// unchanged.
+type Provider struct {


Passthrough stays disabled
Embedding ChatCompatible makes Fireworks implement Passthrough, but the server only exposes /p/{provider}/... for the allowlist in config/config.go and internal/server/passthrough_support.go, which currently omits fireworks. With the documented default setup of only FIREWORKS_API_KEY, /p/fireworks/... returns provider passthrough for "fireworks" is not enabled, so the advertised passthrough surface is not reachable unless users override ENABLED_PASSTHROUGH_PROVIDERS.

mintlify Bot deployed to staging - docs July 4, 2026 16:29 View deployment

Merge branch 'main' into feat/fireworks-provider

20ada98

mintlify Bot deployed to staging - docs July 4, 2026 16:34 View deployment

greptile-apps Bot reviewed Jul 4, 2026

View reviewed changes

SantiagoDePolonia merged commit 49f895e into main Jul 4, 2026
20 checks passed

greptile-apps Bot reviewed Jul 4, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

feat(fireworks): add Fireworks AI provider#475

feat(fireworks): add Fireworks AI provider#475
SantiagoDePolonia merged 2 commits into
mainfrom
feat/fireworks-provider

SantiagoDePolonia commented Jul 4, 2026

Uh oh!

mintlify Bot commented Jul 4, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Jul 4, 2026 •

edited

Loading

Review limit reached

Uh oh!

codecov-commenter commented Jul 4, 2026

Uh oh!

greptile-apps Bot commented Jul 4, 2026

Uh oh!

greptile-apps Bot Jul 4, 2026

Uh oh!

Uh oh!

greptile-apps Bot commented Jul 4, 2026

T-Rex Logs

Uh oh!

greptile-apps Bot Jul 4, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Uh oh!

Conversation

SantiagoDePolonia commented Jul 4, 2026

Summary

User-visible impact

Provider-specific behavior

Testing

Uh oh!

mintlify Bot commented Jul 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

coderabbitai Bot commented Jul 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review limit reached

Uh oh!

codecov-commenter commented Jul 4, 2026

Codecov Report

Uh oh!

greptile-apps Bot commented Jul 4, 2026

Confidence Score: 4/5

Sequence Diagram

Uh oh!

greptile-apps Bot Jul 4, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

greptile-apps Bot commented Jul 4, 2026

Confidence Score: 4/5

T-Rex Logs

Sequence Diagram

Uh oh!

greptile-apps Bot Jul 4, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

mintlify Bot commented Jul 4, 2026 •

edited

Loading

coderabbitai Bot commented Jul 4, 2026 •

edited

Loading