Deprecations

We regularly update our platform with the latest and most powerful open-source models.

Overview

This document outlines our deprecation policy and provides information on migrating from deprecated models to newer or alternate versions.

Deprecation Policy

Model Type	Deprecation Notice	Notes
Preview Model	<24 hrs of notice, after 30 days	Clearly marked in docs and playground with “Preview” tag
Serverless Endpoint	2 or 3 weeks*
On Demand Dedicated Endpoint	2 or 3 weeks*

*Depends on usage and whether there’s an available newer version of the model.

• Users of models scheduled for deprecation will be notified by email.
• All changes will be reflected on this page.
• Each deprecated model will have a specified removal date.
• After the removal date, the model will no longer be queryable via its serverless endpoint but options to migrate will be available as described below.

Migration Options

When a model is deprecated on our serverless platform, users have three options:

1. On-demand Dedicated Endpoint (if supported):

   • Reserved solely for the user, users choose underlying hardware.
   • Charged on a price per minute basis.
   • Endpoints can be dynamically spun up and down.

2. Monthly Reserved Dedicated Endpoint:

   • Reserved solely for the user.
   • Charged on a month-by-month basis.
   • Can be requested via this form.

3. Migrate to a newer serverless model:

   • Switch to an updated model on the serverless platform.

Migration Steps

1. Review the deprecation table below to find your current model.
2. Check if on-demand dedicated endpoints are supported for your model.
3. Decide on your preferred migration option.
4. If choosing a new serverless model, test your application thoroughly with the new model before fully migrating.
5. Update your API calls to use the new model or dedicated endpoint.

Planned Deprecations

Planned Deprecation Date	Model	Recommended Model Replacement
2025-04-17	google-gemma-2-9b-it	Qwen/Qwen2.5-7B-Instruct-Turbo
2025-04-17	google-gemma-2b-it	Qwen/Qwen2.5-7B-Instruct-Turbo
2025-04-17	gryphe-mythomax-l2-13b-lite	gryphe-mythomax-l2-13b
2025-04-17	meta-llama-llama-3-2-3b-instruct-turbo-lora	meta-llama/Meta-Llama-3.1-8B-Instruct-lora
2025-04-30	gryphe-mythomax-l2-13b	meta-llama/Llama-3.3-70B-Instruct-Turbo
2025-04-30	mistralai-mixtral-8x22b-instruct-v0-1	meta-llama/Llama-3.3-70B-Instruct-Turbo
2025-04-30	mistralai-mixtral-8x7b-v0-1	meta-llama/Llama-3.3-70B-Instruct-Turbo
2025-04-30	togethercomputer-m2-bert-80m-2k-retrieval	togethercomputer/m2-bert-80M-32k-retrieval
2025-04-30	togethercomputer-m2-bert-80m-8k-retrieval	togethercomputer/m2-bert-80M-32k-retrieval
2025-04-30	whereisai-uae-large-v1	togethercomputer/m2-bert-80M-32k-retrieval

Deprecation History

All deprecations are listed below, with the most recent deprecations at the top.

Removal Date	Model	Supported by on-demand dedicated endpoints
2025-04-17	meta-llama-meta-llama-3-8b-instruct-turbo	No (coming soon!)
2025-04-24	meta-llama/Llama-2-13b-chat-hf	No (coming soon!)
2025-04-24	xmeta-llama-meta-llama-3-70b-instruct-turbo	No (coming soon!)
2025-04-24	meta-llama-meta-llama-3-1-8b-instruct-turbo-lora	No (coming soon!)
2025-04-24	meta-llama-meta-llama-3-1-70b-instruct-turbo-lora	No (coming soon!)
2025-04-24	meta-llama-llama-3-2-1b-instruct-lora	No (coming soon!)
2025-04-24	microsoft-wizardlm-2-8x22b	No (coming soon!)
2025-04-24	upstage-solar-10-7b-instruct-v1	No (coming soon!)
2025-04-14	stabilityai/stable-diffusion-xl-base-1.0	No (coming soon!)
2025-04-04	meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo-lora	No (coming soon!)
2025-03-27	mistralai/Mistral-7B-v0.1	No
2025-03-25	Qwen/QwQ-32B-Preview	No
2025-03-13	databricks-dbrx-instruct	No
2025-03-11	meta-llama/Meta-Llama-3-70B-Instruct-Lite	No
2025-03-08	Meta-Llama/Llama-Guard-7b	No
2025-02-06	sentence-transformers/msmarco-bert-base-dot-v5	No
2025-02-06	bert-base-uncased	No
2024-10-29	Qwen/Qwen1.5-72B-Chat	No
2024-10-29	Qwen/Qwen1.5-110B-Chat	No
2024-10-07	NousResearch/Nous-Hermes-2-Yi-34B	No
2024-10-07	NousResearch/Hermes-3-Llama-3.1-405B-Turbo	No
2024-08-22	NousResearch/Nous-Hermes-2-Mistral-7B-DPO	Yes
2024-08-22	SG161222/Realistic_Vision_V3.0_VAE	No
2024-08-22	meta-llama/Llama-2-70b-chat-hf	No
2024-08-22	mistralai/Mixtral-8x22B	No
2024-08-22	Phind/Phind-CodeLlama-34B-v2	No
2024-08-22	meta-llama/Meta-Llama-3-70B	Yes
2024-08-22	teknium/OpenHermes-2p5-Mistral-7B	Yes
2024-08-22	openchat/openchat-3.5-1210	Yes
2024-08-22	WizardLM/WizardCoder-Python-34B-V1.0	No
2024-08-22	NousResearch/Nous-Hermes-2-Mixtral-8x7B-SFT	Yes
2024-08-22	NousResearch/Nous-Hermes-Llama2-13b	Yes
2024-08-22	zero-one-ai/Yi-34B-Chat	No
2024-08-22	codellama/CodeLlama-34b-Instruct-hf	No
2024-08-22	codellama/CodeLlama-34b-Python-hf	No
2024-08-22	teknium/OpenHermes-2-Mistral-7B	Yes
2024-08-22	Qwen/Qwen1.5-14B-Chat	Yes
2024-08-22	stabilityai/stable-diffusion-2-1	No
2024-08-22	meta-llama/Llama-3-8b-hf	Yes
2024-08-22	prompthero/openjourney	No
2024-08-22	runwayml/stable-diffusion-v1-5	No
2024-08-22	wavymulder/Analog-Diffusion	No
2024-08-22	Snowflake/snowflake-arctic-instruct	No
2024-08-22	deepseek-ai/deepseek-coder-33b-instruct	No
2024-08-22	Qwen/Qwen1.5-7B-Chat	Yes
2024-08-22	Qwen/Qwen1.5-32B-Chat	No
2024-08-22	cognitivecomputations/dolphin-2.5-mixtral-8x7b	No
2024-08-22	garage-bAInd/Platypus2-70B-instruct	No
2024-08-22	google/gemma-7b-it	Yes
2024-08-22	meta-llama/Llama-2-7b-chat-hf	Yes
2024-08-22	Qwen/Qwen1.5-32B	No
2024-08-22	Open-Orca/Mistral-7B-OpenOrca	Yes
2024-08-22	codellama/CodeLlama-13b-Instruct-hf	Yes
2024-08-22	NousResearch/Nous-Capybara-7B-V1p9	Yes
2024-08-22	lmsys/vicuna-13b-v1.5	Yes
2024-08-22	Undi95/ReMM-SLERP-L2-13B	Yes
2024-08-22	Undi95/Toppy-M-7B	Yes
2024-08-22	meta-llama/Llama-2-13b-hf	No
2024-08-22	codellama/CodeLlama-70b-Instruct-hf	No
2024-08-22	snorkelai/Snorkel-Mistral-PairRM-DPO	Yes
2024-08-22	togethercomputer/LLaMA-2-7B-32K-Instruct	Yes
2024-08-22	Austism/chronos-hermes-13b	Yes
2024-08-22	Qwen/Qwen1.5-72B	No
2024-08-22	zero-one-ai/Yi-34B	No
2024-08-22	codellama/CodeLlama-7b-Instruct-hf	Yes
2024-08-22	togethercomputer/evo-1-131k-base	No
2024-08-22	codellama/CodeLlama-70b-hf	No
2024-08-22	WizardLM/WizardLM-13B-V1.2	Yes
2024-08-22	meta-llama/Llama-2-7b-hf	No
2024-08-22	google/gemma-7b	Yes
2024-08-22	Qwen/Qwen1.5-1.8B-Chat	Yes
2024-08-22	Qwen/Qwen1.5-4B-Chat	Yes
2024-08-22	lmsys/vicuna-7b-v1.5	Yes
2024-08-22	zero-one-ai/Yi-6B	Yes
2024-08-22	Nexusflow/NexusRaven-V2-13B	Yes
2024-08-22	google/gemma-2b	Yes
2024-08-22	Qwen/Qwen1.5-7B	Yes
2024-08-22	NousResearch/Nous-Hermes-llama-2-7b	Yes
2024-08-22	togethercomputer/alpaca-7b	Yes
2024-08-22	Qwen/Qwen1.5-14B	Yes
2024-08-22	codellama/CodeLlama-70b-Python-hf	No
2024-08-22	Qwen/Qwen1.5-4B	Yes
2024-08-22	togethercomputer/StripedHyena-Hessian-7B	No
2024-08-22	allenai/OLMo-7B-Instruct	No
2024-08-22	togethercomputer/RedPajama-INCITE-7B-Instruct	No
2024-08-22	togethercomputer/LLaMA-2-7B-32K	Yes
2024-08-22	togethercomputer/RedPajama-INCITE-7B-Base	No
2024-08-22	Qwen/Qwen1.5-0.5B-Chat	Yes
2024-08-22	microsoft/phi-2	Yes
2024-08-22	Qwen/Qwen1.5-0.5B	Yes
2024-08-22	togethercomputer/RedPajama-INCITE-7B-Chat	No
2024-08-22	togethercomputer/RedPajama-INCITE-Chat-3B-v1	No
2024-08-22	togethercomputer/GPT-JT-Moderation-6B	No
2024-08-22	Qwen/Qwen1.5-1.8B	Yes
2024-08-22	togethercomputer/RedPajama-INCITE-Instruct-3B-v1	No
2024-08-22	togethercomputer/RedPajama-INCITE-Base-3B-v1	No
2024-08-22	WhereIsAI/UAE-Large-V1	No
2024-08-22	allenai/OLMo-7B	No
2024-08-22	togethercomputer/evo-1-8k-base	No
2024-08-22	WizardLM/WizardCoder-15B-V1.0	No
2024-08-22	codellama/CodeLlama-13b-Python-hf	Yes
2024-08-22	allenai-olmo-7b-twin-2t	No
2024-08-22	sentence-transformers/msmarco-bert-base-dot-v5	No
2024-08-22	codellama/CodeLlama-7b-Python-hf	Yes
2024-08-22	hazyresearch/M2-BERT-2k-Retrieval-Encoder-V1	No
2024-08-22	bert-base-uncased	No
2024-08-22	mistralai/Mistral-7B-Instruct-v0.1-json	No
2024-08-22	mistralai/Mistral-7B-Instruct-v0.1-tools	No
2024-08-22	togethercomputer-codellama-34b-instruct-json	No
2024-08-22	togethercomputer-codellama-34b-instruct-tools	No

Notes on model support:

• Models marked “Yes” in the on-demand dedicated endpoint support column can be spun up as dedicated endpoints with customizable hardware.
• Models marked “No” are not available as on-demand endpoints and will require migration to a different model or a monthly reserved dedicated endpoint.

Recommended Actions

• Regularly check this page for updates on model deprecations.
• Plan your migration well in advance of the removal date to ensure a smooth transition.
• If you have any questions or need assistance with migration, please contact our support team.

For the most up-to-date information on model availability, support, and recommended alternatives, please check our API documentation or contact our support team.