Foundation model lifecycle

To help you discover and use the latest and best foundation models, the list of foundation models that are deployed by IBM on multitenant hardware and available for prompting in watsonx.ai is updated regularly.

Foundation models that are built by IBM are continuously updated and improved. As new versions of IBM foundation models are introduced, older versions remain available for you to use for at least 90 days after an updated model is introduced.

Similarly, as newer and more-effective models from third-party providers become available, older models are removed from watsonx.ai. You are given at least 30 days notice before foundation models from third-party providers are removed from watsonx.ai.

Some foundation models that are withdrawn and no longer available for prompting from multitenant hardware can be deployed on demand or provisioned as a custom foundation model. For more information, see Deploy on demand foundation models and Deploying custom foundation models.

Modifications to IBM foundation models

IBM foundation models are periodically modified by IBM to improve the foundation model performance or security. A modification is a model refresh that might include new capabilities or fixes, but does not meet IBM's criteria to warrant a version update.

Foundation model modifications do not disrupt the watsonx.ai service. You can check the current full version number for an IBM foundation model at any time from the model card. The version number consists of three digits that identify the version, modification, and fix numbers that are associated with the IBM foundation model. For more information about versioning, see IBM Software product versioning explained.

Any applications that inference an IBM foundation model that is modified will pick up the modifications, including any changes in performance or in the output that is generated by the model.

Foundation model deprecation

During the deprecation period, you can continue to inference the deprecated foundation model. However, a message is returned with the foundation model output to notify you that the model is deprecated and a new version is available.

A deprecated foundation model can also be constricted. When a deprecated model is in the constricted state, it means the model can be inferenced, but cannot be tuned, trained, or deployed.

When a foundation model is deprecated, the following steps are taken to inform you about the deprecation:

  • The foundation model is highlighted in the product user interface with a warning icon Warning icon. A tooltip indicates that the deprecated model is scheduled for withdrawal.
  • The deprecation is announced in the What’s new topic of the product documentation. The release note clearly states the deprecation date and withdrawal date for the foundation model.
  • The Deprecated foundation models table is updated to show the foundation model that is being deprecated, the dates of deprecation and withdrawal, and a suitable alternative foundation model for you to consider as a replacement.

Deprecated and withdrawn models

The following tables list the foundation models that are deprecated and scheduled for withdrawal from the watsonx platform based on deployment method:

Table 1: Deprecated and withdrawn multitenant foundation models
Foundation model name
API model ID
Availability date Deprecation date Withdrawal date Recommended alternative model
llama-3-405b-instruct
llama-3-405b-instruct
23 July 2024 24 November 2025 29 January 2026 llama-4-maverick-17b-128e-instruct-fp8
llama-3-2-90b-vision-instruct
llama-3-2-90b-vision-instruct
25 September 2025 24 November 2025 29 January 2026 llama-4-maverick-17b-128e-instruct-fp8
granite-3-3-8b-instruct
granite-3-3-8b-instruct
16 April 2025 24 November 2025 22 February 2026 granite-4-h-small
granite-3-2-8b-instruct
granite-3-2-8b-instruct
20 February 2025 24 November 2025 22 February 2026 granite-4-h-small
granite-3-8b-instruct
granite-3-8b-instruct
21 October 2024 24 November 2025 22 February 2026 granite-4-h-small
mistral-large (Toronto data center only)
mistralai/mistral-large
21 June 2024 9 July 2025 12 December 2025 mistral-medium-2505
granite-vision-3-2-2b
ibm/granite-vision-3-2-2b
21 February 2025 13 August 2025 12 November 2025
granite-guardian-3-8b
ibm/granite-guardian-3-8b
(Sydney and Mumbai (AWS) data center)
21 October 2024 13 August 2025 12 November 2025
granite-3-2b-instruct
ibm/granite-3-2b-instruct
21 October 2024 13 August 2025 12 November 2025 granite-3-3-8b-instruct
llama-3-2-3b-instruct
meta-llama/llama-3-2-3b-instruct
25 September 2024 13 August 2025 12 September 2025 llama-4-maverick-17b-128e-instruct-fp8
llama-3-2-1b-instruct
meta-llama/llama-3-2-1b-instruct
25 September 2024 13 August 2025 12 September 2025 llama-4-maverick-17b-128e-instruct-fp8
jais-13b-chat
core42/jais-13b-chat
11 April 2024 13 August 2025 12 September 2025
elyza-japanese-llama-2-7b-instruct
elyza/elyza-japanese-llama-2-7b-instruct
1 January 2024 9 July 2025 10 September 2025 llama-4-maverick-17b-128e-instruct-fp8
granite-guardian-3-2b
ibm/granite-guardian-3-2b
21 October 2024 9 July 2025 8 October 2025 granite-guardian-3-8b
mistral-large (Dallas, Frankfurt, London, Sydney data centers)
mistralai/mistral-large
21 June 2024 9 July 2025 8 October 2025 mistral-medium-2505
pixtral-12b
mistralai/pixtral-12b
21 September 2024 9 July 2025 8 October 2025 mistral-small-3-1-24b-instruct-2503
granite-13b-instruct-v2
ibm/granite-13b-instruct-v2
1 December 2023 19 June 2025 15 October 2025
flan-t5-xl-3b
google/flan-t5-xl
7 December 2023 19 June 2025 15 October 2025
flan-t5-xxl-11b
google/flan-t5-xxl
7 July 2023 28 May 2025 30 July 2025
flan-ul2-20b
google/flan-ul2
7 July 2023 28 May 2025 30 July 2025
llama-4-scout-17b-16e-instruct
meta-llama/llama-4-scout-17b-16e-instruct
7 April 2025 14 May 2025 4 June 2025 llama-4-maverick-17b-128e-instruct-fp8
mistral-small-24b-instruct-2501
mistralai/mistral-small-24b-instruct-2501
30 January 2025 30 April 2025 2 July 2025 mistral-small-3-1-24b-instruct-2503
mixtral-8x7b-instruct-v01
mistralai/mixtral-8x7b-instruct-v01
17 April 2025 30 April 2025 30 July 2025 mistral-small-3-1-24b-instruct-2503
granite-3b-code-instruct
ibm/granite-3b-code-instruct
9 May 2024 16 April 2025 17 July 2025 granite-3-3-8b-instruct
granite-20b-code-instruct
ibm/granite-20b-code-instruct
6 May 2024 16 April 2025 17 July 2025 granite-3-3-8b-instruct
granite-34b-code-instruct
ibm/granite-34b-code-instruct
6 May 2024 16 April 2025 17 July 2025 granite-3-3-8b-instruct
granite-8b-japanese
ibm/granite-8b-japanese
29 February 2024 16 April 2025 20 August 2025 granite-3-8b-instruct
granite-20b-multilingual
ibm/granite-20b-multilingual
14 March 2024
1.1.0: 18 April 2024
15 January 2025 16 April 2025 granite-3-8b-instruct
llama-3-1-8b-instruct
meta-llama/llama-3-1-8b-instruct
1 August 2024 22 January 2025 30 May 2025 llama-3-2-11b-vision-instruct
llama-3-1-70b-instruct
meta-llama/llama-3-1-70b-instruct
7 August 2024 22 January 2025 30 May 2025 llama-3-3-70b-instruct, llama-3-2-90b-vision-instruct
codellama-34b-instruct
codellama/codellama-34b-instruct-hf
14 March 2024 15 January 2025 2 April 2025 llama-3-3-70b-instruct
llama-3-70b-instruct
meta-llama/llama-3-70b-instruct
(London and Sydney data center)
18 April 2024 2 December 2024 2 April 2025 llama-3-3-70b-instruct, llama-3-2-90b-vision-instruct
llama-2-13b-chat
meta-llama/llama-2-13b-chat
11 September 2023 26 August 2024 llama-3-2-11b-vision-instruct
granite-13b-chat-v2
ibm/granite-13b-chat-v2
30 November 2023
2.1.0: 15 February 2024
4 November 2024 3 February 2025 granite-3-8b-instruct
llama-3-8b-instruct
meta-llama/llama-3-8b-instruct
18 April 2024 2 December 2024 3 February 2025 llama-3-2-11b-vision-instruct
llama-3-70b-instruct
meta-llama/llama-3-70b-instruct
(Dallas, Frankfurt, and Tokyo data centers)
18 April 2024 2 December 2024 3 February 2025 llama-3-3-70b-instruct, llama-3-2-90b-vision-instruct
granite-7b-lab
ibm/granite-7b-lab
7 May 2024 7 October 2024 7 January 2025 granite-3-8b-instruct
llama2-13b-dpo-v7
mnci/llama2-13b-dpo-v7
18 April 2024 4 November 2024 4 December 2024 llama-3-1-8b-instruct
mt0-xxl-13b
bigscience/mt0-xxl
7 July 2023 4 November 2024 4 December 2024 llama-3-1-8b-instruct, llama-3-2-11b-vision-instruct
llama3-llava-next-8b-hf
meta-llama/llama3-llava-next-8b-hf
19 September 2024 7 October 2024 7 November 2024 llama-3-2-11b-vision-instruct
llama-2-70b-chat
meta-llama/llama-2-70b-chat
11 September 2023 26 August 2024 25 September 2024 llama-3.1-70b-instruct
mixtral-8x7b-instruct-v01-q
ibm-mistralai/mixtral-8x7b-instruct-v01-q
15 February 2024 19 April 2024 30 August 2024 mixtral-8x7b-instruct-v01
merlinite-7b
ibm-mistralai/merlinite-7b
7 May 2024 22 July 2024 22 August 2024 mixtral-8x7b-instruct-v01
starcoder-15.5b
bigcode/starcoder
31 August 2023 15 February 2024 25 April 2024 codellama-34b-instruct
mpt-7b-instruct2
ibm/mpt-7b-instruct2
7 July 2023 15 February 2024 21 March 2024 mixtral-8x7b-instruct-v01-q
gpt-neox-20b
eleutherai/gpt-neox-20b
7 July 2023 15 February 2024 21 March 2024 mixtral-8x7b-instruct-v01-q
granite-13b-chat-v1
ibm/granite-13b-chat-v1
28 September 2023 11 January 2024 11 April 2024 granite-13b-chat-v2
granite-13b-instruct-v1
ibm/granite-13b-instruct-v1
28 September 2023 11 January 2024 11 April 2024 granite-13b-instruct-v2
Table 2: Deprecated and withdrawn deploy on demand foundation models
Foundation model name
API model ID
Availability date Deprecation date Withdrawal date
granite-13b-instruct-v2
ibm/granite-13b-instruct-v2
1 December 2023 19 June 2025 15 October 2025
flan-t5-xl-3b
google/flan-t5-xl
7 December 2023 19 June 2025 15 October 2025
flan-t5-xxl-11b
google/flan-t5-xxl
7 July 2023 28 May 2025 30 July 2025
flan-ul2-20b
google/flan-ul2
7 July 2023 28 May 2025 30 July 2025
Table 3: Deprecated and withdrawn embedding models
Foundation model name
API model ID
Availability date Deprecation date Withdrawal date Recommended alternative model
slate-125m-english-rtrvr
ibm/slate-125m-english-rtrvr
11 April 2024 13 August 2025 12 November 2025 slate-125m-english-rtrvr-v2
slate-30m-english-rtrvr
ibm/slate-30m-english-rtrvr
9 August 2024 13 August 2025 12 November 2025 slate-30m-english-rtrvr-v2
granite-embedding-107m-multilingual
ibm/granite-embedding-107m-multilingual
6 January 2025 13 August 2025 12 November 2025 granite-278m-multilingual-embedding
all-minilm-l12-v2
sentence-transformers/all-minilm-l12-v2
3 May 2024 13 August 2025 12 September 2025 all-minilm-l6-v2

What to do next

You must choose an alternative supported foundation model to use if any of the following saved resources submit input to a foundation model that is withdrawn:

  • Prompt template asset
  • Prompt session asset
  • Notebook asset

For details about working with saved prompt assets, see Saving your work.

For details about how to change the foundation model that is inferenced from a notebook asset, see Inferencing a foundation model with a notebook.