Skip to content

Commit 451779f

Browse files
devin-ai-integration[bot]devin-ai-integration[bot]
committed
Merge remote-tracking branch 'origin/main' into devin/1764806701-keyterms-multilingual-support
2 parents 31dd37b + 07effa5 commit 451779f

File tree

7 files changed

+76
-23
lines changed

7 files changed

+76
-23
lines changed

‎fern/.definition/__package__.yml‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -52,3 +52,10 @@ errors:
5252
error: Authentication error, API token missing/invalid
5353
status-code: 401
5454
type: Error
55+
CannotAccessUploadedFileError:
56+
docs: Cannot access uploaded file
57+
examples:
58+
- value:
59+
error: Cannot access uploaded file
60+
status-code: 403
61+
type: Error

‎fern/.definition/files.yml‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -21,10 +21,13 @@ service:
2121
<Note>To upload a media file to our EU server, replace
2222
`api.assemblyai.com` with `api.eu.assemblyai.com`.</Note>
2323
24+
<Warning>Requests to transcribe uploaded files must use an API key from the same project as the key that was used to upload the file. If you use an API key from a different project you will get a `403` error and "Cannot access uploaded file" message.</Warning>
25+
2426
Upload a media file to AssemblyAI's servers.
2527
errors:
2628
- root.BadRequestError
2729
- root.UnauthorizedError
30+
- root.CannotAccessUploadedFileError
2831
- root.NotFoundError
2932
- root.TooManyRequestsError
3033
- root.InternalServerError

‎fern/docs.yml‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -972,6 +972,9 @@ navigation:
972972
- page: Do you offer EU Data Residency?
973973
path: pages/faq/privacy-security/do-you-offer-eu-data-residency.mdx
974974
slug: /do-you-offer-eu-data-residency
975+
- page: Do you offer self-hosted solutions?
976+
path: pages/faq/privacy-security/do-you-offer-self-hosted-solutions.mdx
977+
slug: /do-you-offer-self-hosted-solutions
975978
- page: Do you offer servers in the EU?
976979
path: pages/faq/privacy-security/do-you-offer-servers-in-the-eu.mdx
977980
slug: /do-you-offer-servers-in-the-eu

‎fern/pages/07-llm-gateway/llm-gateway.mdx‎

Lines changed: 26 additions & 22 deletions
Original file line numberDiff line numberDiff line change
@@ -20,34 +20,38 @@ The LLM Gateway provides access to 15+ models across major AI providers with sup
2020

2121
### Anthropic Claude
2222

23-
| Model | Parameter | Latency per 10,000 tokens | [LMArena Score](https://lmarena.ai/leaderboard) | Description |
24-
| --------------------- | ---------------------------- | ------------------------- | ------------- | ------------------------------------------------------ |
25-
| **Claude 4.5 Sonnet** | `claude-sonnet-4-5-20250929` | 10.1s | 1444 | Claude's best model for complex agents and coding |
26-
| **Claude 4 Sonnet** | `claude-sonnet-4-20250514` | 7.1s | 1389 | High-performance model |
27-
| **Claude 4 Opus** | `claude-opus-4-20250514` | 15.4s | 1412 | Claude's previous flagship model |
28-
| **Claude 4.5 Haiku** | `claude-haiku-4-5-20251001` | 4.6s | 1402 | Claude's fastest and most intelligent Haiku model |
29-
| **Claude 3.5 Haiku** | `claude-3-5-haiku-20241022` | 5.4s | 1322 | Fast and efficient model with strong performance |
30-
| **Claude 3.0 Haiku** | `claude-3-haiku-20240307` | 4.8s | 1262 | Fast and compact model for near-instant responsiveness |
23+
| Model | Parameter | Latency per 10,000 tokens | [LMArena Score](https://lmarena.ai/leaderboard) | Description | Retention Policy | Anthropic Model Training |
24+
| --------------------- | ---------------------------- | ------------------------- | ------------- | ------------------------------------------------------ | ---------------- | ------------------------ |
25+
| **Claude 4.5 Sonnet** | `claude-sonnet-4-5-20250929` | 10.1s | 1444 | Claude's best model for complex agents and coding | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies.* | AssemblyAI has opted out of model training with all LLM Gateway providers. |
26+
| **Claude 4 Sonnet** | `claude-sonnet-4-20250514` | 7.1s | 1389 | High-performance model | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies.* | AssemblyAI has opted out of model training with all LLM Gateway providers. |
27+
| **Claude 4 Opus** | `claude-opus-4-20250514` | 15.4s | 1412 | Claude's previous flagship model | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies.* | AssemblyAI has opted out of model training with all LLM Gateway providers. |
28+
| **Claude 4.5 Haiku** | `claude-haiku-4-5-20251001` | 4.6s | 1402 | Claude's fastest and most intelligent Haiku model | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies.* | AssemblyAI has opted out of model training with all LLM Gateway providers. |
29+
| **Claude 3.5 Haiku** | `claude-3-5-haiku-20241022` | 5.4s | 1322 | Fast and efficient model with strong performance | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies.* | AssemblyAI has opted out of model training with all LLM Gateway providers. |
30+
| **Claude 3.0 Haiku** | `claude-3-haiku-20240307` | 4.8s | 1262 | Fast and compact model for near-instant responsiveness | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies.* | AssemblyAI has opted out of model training with all LLM Gateway providers. |
31+
32+
<Note>
33+
*If Amazon Bedrock fails, for non-EU customers we may send your request to the Anthropic API, where we have 0-day retention configured. Please see Anthropic's commercial terms [here](https://www.anthropic.com/legal/commercial-terms).
34+
</Note>
3135

3236
### OpenAI GPT
3337

34-
| Model | Parameter | Latency per 10,000 tokens | [LMArena Score](https://lmarena.ai/leaderboard) | Description |
35-
| ---------------- | ------------------- | ------------------------- | ------------- | ---------------------------------------------------------------- |
36-
| **GPT-5** | `gpt-5` | 18.9s | 1425 | OpenAI's best model for coding and agentic tasks across domains |
37-
| **GPT-5 nano** | `gpt-5-nano` | 11.2s | 1338 | OpenAI's fastest, most cost-efficient version of GPT-5 |
38-
| **GPT-5 mini** | `gpt-5-mini` | 21.9s | 1393 | A faster, cost-efficient version of GPT-5 for well-defined tasks |
39-
| **GPT-4.1** | `gpt-4.1` | 12.6s | 1412 | OpenAI's smartest non-reasoning model |
40-
| **gpt-oss-120b** | `gpt-oss-120b` | 10.5s | 1352 | OpenAI's most powerful open-weight model |
41-
| **gpt-oss-20b** | `gpt-oss-20b` | 4.2s | 1318 | Medium-sized open-weight model for low latency |
38+
| Model | Parameter | Latency per 10,000 tokens | [LMArena Score](https://lmarena.ai/leaderboard) | Description | Retention Policy | OpenAI Model Training |
39+
| ---------------- | ------------------- | ------------------------- | ------------- | ---------------------------------------------------------------- | ---------------- | --------------------- |
40+
| **GPT-5** | `gpt-5` | 18.9s | 1425 | OpenAI's best model for coding and agentic tasks across domains | Abuse monitoring retains logs for 30 days. If you require ZDR, please use Anthropic or Google models. | AssemblyAI has opted out of model training with all LLM Gateway providers. |
41+
| **GPT-5 nano** | `gpt-5-nano` | 11.2s | 1338 | OpenAI's fastest, most cost-efficient version of GPT-5 | Abuse monitoring retains logs for 30 days. If you require ZDR, please use Anthropic or Google models. | AssemblyAI has opted out of model training with all LLM Gateway providers. |
42+
| **GPT-5 mini** | `gpt-5-mini` | 21.9s | 1393 | A faster, cost-efficient version of GPT-5 for well-defined tasks | Abuse monitoring retains logs for 30 days. If you require ZDR, please use Anthropic or Google models. | AssemblyAI has opted out of model training with all LLM Gateway providers. |
43+
| **GPT-4.1** | `gpt-4.1` | 12.6s | 1412 | OpenAI's smartest non-reasoning model | Abuse monitoring retains logs for 30 days. If you require ZDR, please use Anthropic or Google models. | AssemblyAI has opted out of model training with all LLM Gateway providers. |
44+
| **gpt-oss-120b** | `gpt-oss-120b` | 10.5s | 1352 | OpenAI's most powerful open-weight model | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies. | AssemblyAI has opted out of model training with all LLM Gateway providers. |
45+
| **gpt-oss-20b** | `gpt-oss-20b` | 4.2s | 1318 | Medium-sized open-weight model for low latency | We use this model through Amazon Bedrock. Amazon Bedrock doesn't store or log your prompts and completions. Amazon Bedrock doesn't use your prompts and completions to train any AWS models and doesn't distribute them to third parties. See [here](https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html) for more information on Amazon Bedrock data protection policies. | AssemblyAI has opted out of model training with all LLM Gateway providers. |
4246

4347
### Google Gemini
4448

45-
| Model | Parameter | Latency per 10,000 tokens | [LMArena Score](https://lmarena.ai/leaderboard) | Description |
46-
| ------------------------- | ----------------------- | ------------------------- | ------------- | ------------------------------------------------------------------------------------- |
47-
| **Gemini 3 Pro Preview** | `gemini-3-pro-preview` | TBD | 1495 | Gemini's most powerful agentic and vibe-coding model, delivering richer visuals and deeper interactivity |
48-
| **Gemini 2.5 Pro** | `gemini-2.5-pro` | 13.9s | 1451 | Gemini's state-of-the-art thinking model, capable of reasoning over complex problems |
49-
| **Gemini 2.5 Flash** | `gemini-2.5-flash` | 8.3s | 1407 | Gemini's best model in terms of price-performance, offering well-rounded capabilities |
50-
| **Gemini 2.5 Flash-Lite** | `gemini-2.5-flash-lite` | 1.6s | 1375 | Gemini's fastest flash model optimized for cost-efficiency and high throughput |
49+
| Model | Parameter | Latency per 10,000 tokens | [LMArena Score](https://lmarena.ai/leaderboard) | Description | Retention Policy | Google Model Training |
50+
| ------------------------- | ----------------------- | ------------------------- | ------------- | -------------------------------------------------------------------------------------------------------- | ---------------- | -------------- |
51+
| **Gemini 3 Pro Preview** | `gemini-3-pro-preview` | TBD | 1495 | Gemini's most powerful agentic and vibe-coding model, delivering richer visuals and deeper interactivity | ZDR (see Google's policy [here](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/vertex-ai-zero-data-retention) for more information on how Google defines ZDR) | AssemblyAI has opted out of model training with all LLM Gateway providers. |
52+
| **Gemini 2.5 Pro** | `gemini-2.5-pro` | 13.9s | 1451 | Gemini's state-of-the-art thinking model, capable of reasoning over complex problems | ZDR (see Google's policy [here](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/vertex-ai-zero-data-retention) for more information on how Google defines ZDR) | AssemblyAI has opted out of model training with all LLM Gateway providers. |
53+
| **Gemini 2.5 Flash** | `gemini-2.5-flash` | 8.3s | 1407 | Gemini's best model in terms of price-performance, offering well-rounded capabilities | ZDR (see Google's policy [here](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/vertex-ai-zero-data-retention) for more information on how Google defines ZDR) | AssemblyAI has opted out of model training with all LLM Gateway providers. |
54+
| **Gemini 2.5 Flash-Lite** | `gemini-2.5-flash-lite` | 1.6s | 1375 | Gemini's fastest flash model optimized for cost-efficiency and high throughput | ZDR (see Google's policy [here](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/vertex-ai-zero-data-retention) for more information on how Google defines ZDR) | AssemblyAI has opted out of model training with all LLM Gateway providers. |
5155

5256
Unsure which model to choose?
5357

‎fern/pages/faq/privacy-security/do-you-offer-self-hosted-solutions.mdx‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
---
2+
title: "Do you offer self-hosted solutions?"
3+
---
4+
5+
Yes, we offer self-hosted solutions for organizations that require greater control over their data and infrastructure. Our self-hosted options allow you to deploy our services within your own environment, ensuring compliance with your organization's security policies and regulatory requirements.
6+
7+
[Contact our sales team](https://www.assemblyai.com/contact/sales) to discuss your specific needs and to learn more about our self-hosted offerings.

‎openapi.json‎

Lines changed: 17 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -83,6 +83,9 @@
8383
"401": {
8484
"$ref": "#/components/responses/Unauthorized"
8585
},
86+
"403": {
87+
"$ref": "#/components/responses/CannotAccessUploadedFile"
88+
},
8689
"404": {
8790
"$ref": "#/components/responses/NotFound"
8891
},
@@ -6802,6 +6805,20 @@
68026805
}
68036806
}
68046807
},
6808+
"CannotAccessUploadedFile": {
6809+
"x-label": "Cannot access uploaded file",
6810+
"description": "Cannot access uploaded file",
6811+
"content": {
6812+
"application/json": {
6813+
"schema": {
6814+
"$ref": "#/components/schemas/Error"
6815+
},
6816+
"example": {
6817+
"error": "Cannot access uploaded file"
6818+
}
6819+
}
6820+
}
6821+
},
68056822
"NotFound": {
68066823
"x-label": "Not found",
68076824
"description": "Not found",

‎openapi.yml‎

Lines changed: 13 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -39,8 +39,10 @@ paths:
3939
- transcript
4040
summary: Upload a media file
4141
description: |
42-
<Note>To upload a media file to our EU server, replace `api.assemblyai.com` with `api.eu.assemblyai.com`.</Note>
4342
Upload a media file to AssemblyAI's servers.
43+
44+
<Note>To upload a media file to our EU server, replace `api.assemblyai.com` with `api.eu.assemblyai.com`.</Note>
45+
<Warning>Requests to transcribe uploaded files must use an API key from the same project as the key that was used to upload the file. If you use an API key from a different project you will get a `403` error and "Cannot access uploaded file" message.</Warning>
4446
operationId: uploadFile
4547
x-fern-sdk-group-name: files
4648
x-fern-sdk-method-name: upload
@@ -62,6 +64,8 @@ paths:
6264
$ref: "#/components/responses/BadRequest"
6365
"401":
6466
$ref: "#/components/responses/Unauthorized"
67+
"403":
68+
$ref: "#/components/responses/CannotAccessUploadedFile"
6569
"404":
6670
$ref: "#/components/responses/NotFound"
6771
"429":
@@ -6276,6 +6280,14 @@ components:
62766280
$ref: "#/components/schemas/Error"
62776281
example:
62786282
{ "error": "Authentication error, API token missing/invalid" }
6283+
CannotAccessUploadedFile:
6284+
x-label: Cannot access uploaded file
6285+
description: Cannot access uploaded file
6286+
content:
6287+
application/json:
6288+
schema:
6289+
$ref: "#/components/schemas/Error"
6290+
example: { "error": "Cannot access uploaded file" }
62796291
NotFound:
62806292
x-label: Not found
62816293
description: Not found

0 commit comments

Comments
 (0)