Inference APIs

404 model_not_found: llama-4-scout-17b-16e-instruct (Groq)

Llama 4 Scout was the multimodal model Groq itself pointed people to when it retired the Llama 3.2 vision previews in April 2025, so 404 The model `meta-llama/llama-4-scout-17b-16e-instruct` does not exist or you do not have access to it lands on apps that already migrated once. Groq announced the retirement on 17 June 2026 and shut the model down for free and developer-tier accounts on 17 July 2026, the same day as qwen/qwen3-32b. Its sibling llama-4-maverick-17b-128e-instruct had gone in March.

Last verified November 18, 2026 against Groq deprecations, Groq model list · 5 min read

ProviderGroq
HTTP status404 Not Found
Error codemodel_not_found
MessageThe model `meta-llama/llama-4-scout-17b-16e-instruct` does not exist or you do not have access to it.
Announced / shut down17 June 2026 / 17 July 2026
When it happensSince 2026-07-17, on any free or developer-tier request that names this model
Can you wait it out?No — the model was retired; the request fails until you change the model or the provider
Short answer
  • Fastest: change the model string to openai/gpt-oss-120b on Groq. Different model, so re-check your prompts.
  • Least code change: keep meta-llama/llama-4-scout-17b-16e-instruct and change only the base URL; Inference APIs accepts the id and answers with Qwen3-VL 235B, images included.
  • So it does not happen again: read the model id from configuration and add a fallback for 404 and 429.

What happened

Groq's deprecations page lists the notice: emailed 17 June 2026, shut down 17 July 2026, free and developer tiers only; enterprise contracts were not affected. The recommended replacements were openai/gpt-oss-120b or qwen/qwen3.6-27b, both text-only on Groq, and the Qwen one was itself retired on 14 September 2026 in favour of qwen/qwen3.8-27b. Scout had a 128K window and took images; neither replacement takes images, which is the part most migrations trip on.

First, find where the id is pinned

The id is often in more places than the one that errored: a default in a settings file, an environment variable on the server, a second call for summarisation or titles, a framework default.

Find every copy of the id
grep -rn -E "llama-4-scout-17b-16e-instruct" . --include='*.py' --include='*.ts' --include='*.js' --include='*.json' --include='*.yaml' --include='*.yml' --include='*.toml' --include='*.env*'
# and the places grep cannot see: environment variables on the server, a hosted secrets store, the provider dashboard's default model

Option 1 — switch to Groq's recommended replacement

If your requests were text only, change the string to openai/gpt-oss-120b and you stay on Groq's free tier. What is different:

  • It reasons before answering, so raise max_tokens or send reasoning_effort: "low"; a budget sized for Scout's direct answers can come back empty.
  • No images. A request with an image_url part is rejected. Groq's current list has qwen/qwen3.8-27b, which Groq describes as multimodal; if you want to stay on Groq with images, test that id rather than the retired ones.
  • Style and formatting differ; re-run your test set.
One string changes
resp = client.chat.completions.create(
    model="openai/gpt-oss-120b",   # was "meta-llama/llama-4-scout-17b-16e-instruct"
    max_tokens=800,                 # reasoning counts against it
    messages=[{"role": "user", "content": "Hello"}],
)

Option 2 — keep the old id, change the base URL

Inference APIs accepts meta-llama/llama-4-scout-17b-16e-instruct and llama-4-scout-17b-16e-instruct as aliases and serves them with Qwen3-VL 235B, an open vision-language model: images stay in the same image_url message parts, tool calling works, there is no thinking phase to budget for, and the window is 262,144 tokens, at $0.26 / $1.15 per million tokens.

Base URL and key change, model string stays
from openai import OpenAI

client = OpenAI(
    base_url="https://api.inferenceapis.com/v1",   # was https://api.groq.com/openai/v1
    api_key=os.environ["INFERENCE_API_KEY"],
)
resp = client.chat.completions.create(
    model="meta-llama/llama-4-scout-17b-16e-instruct",   # the retired Groq id, accepted as an alias
    messages=[{"role": "user", "content": "Hello"}],
)

Qwen3-VL is a different model from Llama 4 and reads images differently; OCR and document-extraction prompts deserve a re-run on your own samples. For text-only traffic, openai/gpt-oss-120b is served here as well at $0.20 / $0.80. Prices and capabilities: Qwen3-VL 235B here.

Option 3 — stop pinning one provider and one id

Put the model id in configuration and let a 404 or 429 fall through to a second provider. A retirement then degrades to a log line instead of an outage, and you choose when to migrate. LiteLLM's router does this in a few lines; the same shape exists in most frameworks.

LiteLLM proxy: Groq first, fall back on 404/429
# LiteLLM proxy config: Groq first, Inference APIs when Groq answers 404 or 429
model_list:
  - model_name: chat
    litellm_params:
      model: groq/openai/gpt-oss-120b
      api_key: os.environ/GROQ_API_KEY
  - model_name: chat-fallback
    litellm_params:
      model: inferenceapis/openai/gpt-oss-120b
      api_key: os.environ/INFERENCEAPIS_API_KEY

router_settings:
  fallbacks: [{ "chat": ["chat-fallback"] }]

The deprecations tracker lists every Groq retirement with its date and replacement, so you can see the next one coming.

Frequently asked questions

I migrated from llama-3.2-11b-vision-preview to Scout in 2025. Where do I go now?

To a host that still serves a vision model. On Groq, test qwen/qwen3.8-27b, which Groq lists as multimodal. On Inference APIs the Scout id itself keeps working, served by Qwen3-VL 235B, so the second migration is a base-URL change.

Will my image requests fail on gpt-oss-120b?

Yes. GPT-OSS is text-only on every host, including Groq. Keep a vision-capable model for those requests and route text-only requests wherever you like.

Is Maverick still on Groq?

No. meta-llama/llama-4-maverick-17b-128e-instruct was retired for all self-serve tiers on 9 March 2026; it has its own page, and the same alias applies here.

Where Inference APIs fits

Option 2 above keeps the old id working with a base-URL change: Inference APIs accepts meta-llama/llama-4-scout-17b-16e-instruct (and the short form) and serves it with Qwen3-VL 235B, so image parts keep working, at $0.26 / $1.15 per million tokens with no daily cap. For text-only traffic Groq's replacement openai/gpt-oss-120b is served here too at $0.20 / $0.80.

Something changed or wrong? Tell us and we will re-verify the entry.