diff --git a/README.md b/README.md index e2388c4..a65b0c6 100644 --- a/README.md +++ b/README.md @@ -11,10 +11,44 @@ This lists various services that provide free access or credits towards API-base > [!WARNING] > This list explicitly excludes any services that are not legitimate (eg reverse engineers an existing chatbot) +- [Free Providers](#free-providers) + - [OpenRouter](#openrouter) + - [Google AI Studio](#google-ai-studio) + - [NVIDIA NIM](#nvidia-nim) + - [Mistral (La Plateforme)](#mistral-la-plateforme) + - [Mistral (Codestral)](#mistral-codestral) + - [HuggingFace Inference Providers](#huggingface-inference-providers) + - [Cerebras](#cerebras) + - [Groq](#groq) + - [OVH AI Endpoints (Free Beta)](#ovh-ai-endpoints-free-beta) + - [Together (Free)](#together-free) + - [Cohere](#cohere) + - [GitHub Models](#github-models) + - [Chutes](#chutes) + - [Cloudflare Workers AI](#cloudflare-workers-ai) + - [Google Cloud Vertex AI](#google-cloud-vertex-ai) +- [Providers with trial credits](#providers-with-trial-credits) + - [Together](#together) + - [Fireworks](#fireworks) + - [Unify](#unify) + - [Baseten](#baseten) + - [Nebius](#nebius) + - [Novita](#novita) + - [AI21](#ai21) + - [Upstage](#upstage) + - [NLP Cloud](#nlp-cloud) + - [Alibaba Cloud (International) Model Studio](#alibaba-cloud-international-model-studio) + - [Modal](#modal) + - [Hyperbolic](#hyperbolic) + - [SambaNova Cloud](#sambanova-cloud) + - [Scaleway Generative APIs](#scaleway-generative-apis) + ## Free Providers ### [OpenRouter](https://openrouter.ai) +**Limits:** + [20 requests/minute
50 requests/day
1000 requests/day with $10 credit balance](https://openrouter.ai/docs/api-reference/limits) Models share a common quota. @@ -98,27 +132,39 @@ Data is used for training when used outside of the UK/CH/EEA/EU. ### [NVIDIA NIM](https://build.nvidia.com/explore/discover) -Phone number verification required. Models tend to be context window limited. +Phone number verification required. +Models tend to be context window limited. -- [Various open models](https://build.nvidia.com/models) (40 requests/minute) +**Limits:** 40 requests/minute + +- [Various open models](https://build.nvidia.com/models) ### [Mistral (La Plateforme)](https://console.mistral.ai/) -Free tier (Experiment plan) requires opting into data training, requires phone number verification. +* Free tier (Experiment plan) requires opting into data training +* Requires phone number verification. -- [Open and Proprietary Mistral models](https://docs.mistral.ai/getting-started/models/models_overview/) (1 request/second, 500,000 tokens/minute, 1,000,000,000 tokens/month) +**Limits (per-model):** 1 request/second, 500,000 tokens/minute, 1,000,000,000 tokens/month + +- [Open and Proprietary Mistral models](https://docs.mistral.ai/getting-started/models/models_overview/) ### [Mistral (Codestral)](https://codestral.mistral.ai/) -Currently free to use, monthly subscription based, requires phone number verification. +* Currently free to use +* Monthly subscription based +* Requires phone number verification -- Codestral (30 requests/minute, 2,000 requests/day) +**Limits:** 30 requests/minute, 2,000 requests/day -### [HuggingFace Serverless Inference](https://huggingface.co/docs/api-inference/en/index) +- Codestral -Limited to models smaller than 10GB. Some popular models are supported even if they exceed 10GB. +### [HuggingFace Inference Providers](https://huggingface.co/docs/inference-providers/en/index) -- Various open models ([Variable credits per month, currently $0.10](https://huggingface.co/docs/api-inference/pricing)) +HuggingFace Serverless Inference limited to models smaller than 10GB. Some popular models are supported even if they exceed 10GB. + +**Limits:** [$0.10/month in credits](https://huggingface.co/docs/inference-providers/en/pricing) + +- Various open models across supported providers ### [Cerebras](https://cloud.cerebras.ai/) @@ -168,9 +214,9 @@ Free tier restricted to 8K context. Qwen2.5 Coder 32B Instruct12 requests/minute -### [Together](https://together.ai) +### [Together (Free)](https://together.ai) -Up to 60 requests/minute. +**Limits:** Up to 60 requests/minute - [Llama 3.2 11B Vision Instruct](https://together.ai/llama-3-2-11b-free) - [Llama 3.3 70B Instruct](https://together.ai/llama-3-3-70b-free) @@ -178,6 +224,8 @@ Up to 60 requests/minute. ### [Cohere](https://cohere.com) +**Limits:** + [20 requests/minute
1,000 requests/month](https://docs.cohere.com/docs/rate-limits) Models share a common quota. @@ -193,15 +241,19 @@ Models share a common quota. ### [GitHub Models](https://github.com/marketplace/models) -Extremely restrictive input/output token limits. [Rate limits dependent on Copilot subscription tier (Free/Pro/Business/Enterprise)](https://docs.github.com/en/github-models/prototyping-with-ai-models#rate-limits) +Extremely restrictive input/output token limits. + +**Limits:** [Dependent on Copilot subscription tier (Free/Pro/Pro+/Business/Enterprise)](https://docs.github.com/en/github-models/prototyping-with-ai-models#rate-limits) - AI21 Jamba 1.5 Large - AI21 Jamba 1.5 Mini - Codestral 25.01 +- Cohere Command A - Cohere Command R - Cohere Command R 08-2024 - Cohere Command R+ - Cohere Command R+ 08-2024 +- Cohere Embed 4 - Cohere Embed v3 English - Cohere Embed v3 Multilingual - DeepSeek-R1 @@ -218,7 +270,6 @@ Extremely restrictive input/output token limits. [Rate limits dependent on Copil - Meta-Llama-3.1-70B-Instruct - Meta-Llama-3.1-8B-Instruct - Ministral 3B -- Mistral Large - Mistral Large (2407) - Mistral Large 24.11 - Mistral Nemo @@ -250,7 +301,8 @@ Extremely restrictive input/output token limits. [Rate limits dependent on Copil ### [Chutes](https://chutes.ai/) -Distributed, decentralized crypto-based compute. Data is sent to individual hosts. +Distributed, decentralized crypto-based compute. +Data is sent to individual hosts. - DeepCoder 14B Preview - DeepHermes 3 Llama 3 8B Preview @@ -280,7 +332,7 @@ Distributed, decentralized crypto-based compute. Data is sent to individual host ### [Cloudflare Workers AI](https://developers.cloudflare.com/workers-ai) -[10,000 neurons/day](https://developers.cloudflare.com/workers-ai/platform/pricing/#free-allocation) +**Limits:** [10,000 neurons/day](https://developers.cloudflare.com/workers-ai/platform/pricing/#free-allocation) - DeepSeek R1 Distill Qwen 32B - Deepseek Coder 6.7B Base (AWQ) @@ -449,8 +501,8 @@ Very stringent payment verification for Google Cloud. **Credits:** $5 for 3 months **Models:** -- E5-Mistral-7B-Instruct - Llama 3.1 405B +- Llama 3.1 70B - Llama 3.1 8B - Llama 3.2 11B Vision - Llama 3.2 1B @@ -463,10 +515,14 @@ Very stringent payment verification for Google Cloud. - Llama-Guard-3-8B - Qwen/QwQ-32B - Qwen/QwQ-32B-Preview -- Qwen/Qwen2-Audio-7B-Instruct +- Qwen/Qwen2.5-72B-Instruct +- Qwen/Qwen2.5-Coder-32B-Instruct +- allenai/Llama-3.1-Tulu-3-405B - deepseek-ai/DeepSeek-R1 - deepseek-ai/DeepSeek-R1-Distill-Llama-70B - deepseek-ai/DeepSeek-V3-0324 +- tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.3 +- tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.3 ### [Scaleway Generative APIs](https://console.scaleway.com/generative-api/models) diff --git a/src/README_template.md b/src/README_template.md index 17b7dcb..4e41c16 100644 --- a/src/README_template.md +++ b/src/README_template.md @@ -8,6 +8,8 @@ This lists various services that provide free access or credits towards API-base > [!WARNING] > This list explicitly excludes any services that are not legitimate (eg reverse engineers an existing chatbot) +{{TOC}} + ## Free Providers {{MODEL_LIST}} diff --git a/src/pull_available_models.py b/src/pull_available_models.py index ce67324..1447841 100644 --- a/src/pull_available_models.py +++ b/src/pull_available_models.py @@ -11,6 +11,7 @@ from google.cloud import cloudquotas_v1 from mistralai import Mistral from concurrent.futures import ThreadPoolExecutor import time +import re from data import ( MODEL_TO_NAME_MAPPING, @@ -412,56 +413,6 @@ def fetch_scaleway_models(logger): } ) ret_models = sorted(ret_models, key=lambda x: x["name"]) - logger.info("Fetching Scaleway rate limits") - r = requests.get( - "https://www.scaleway.com/en/docs/generative-apis/reference-content/rate-limits/" - ) - r.raise_for_status() - # Extract
content - soup = BeautifulSoup(r.text, "html.parser") - body = str(soup.find("main")) - prompt = f""" -Here is the web page to extract data from: -```html -{body} -``` - """ - logger.info("Extracting model rate limits from the provided web page...") - chat_response = rate_limited_mistral_chat( - mistral_client, - model="mistral-large-latest", - messages=[ - { - "role": "system", - "content": """Extract the model rate limits as only integers from the provided web page into JSON of the following format: -```json -{ - "model name here": { - "requests/minute": 10, - "tokens/minute": 100 - } -} -``` -ONLY OUTPUT JSON!""", - }, - { - "role": "user", - "content": prompt, - }, - ], - response_format={ - "type": "json_object", - }, - temperature=0, - ) - logger.info(chat_response.choices[0].message.content) - extracted_data = json.loads(chat_response.choices[0].message.content) - for model in extracted_data: - for m in ret_models: - if m["id"] == model: - m["limits"] = extracted_data[model] - break - return ret_models @@ -496,11 +447,29 @@ def fetch_chutes_models(logger): return sorted(free_models, key=lambda x: x["name"]) -def get_human_limits(model): +def get_human_limits(model, seperator="
"): if "limits" not in model: return "" limits = model["limits"] - return "
".join([f"{value:,} {key}" for key, value in limits.items()]) + return seperator.join([f"{value:,} {key}" for key, value in limits.items()]) + + +def generate_toc(markdown): + toc_lines = [] + # Find all ## and ### headings, but skip the main title (# ...) + headings = re.findall(r"^(#{2,3}) +(.+)", markdown, re.MULTILINE) + for hashes, title in headings: + # Remove markdown links for anchor text, keep display text + display = re.sub(r"\[(.*?)\]\([^)]*\)", r"\1", title) + # Build anchor (GitHub style) + anchor = display.lower() + anchor = re.sub(r"[^a-z0-9 \-_]", "", anchor) + anchor = anchor.replace(" ", "-") + anchor = anchor.replace("--", "-") + anchor = anchor.strip("-") + indent = " " if len(hashes) == 3 else "" + toc_lines.append(f"{indent}- [{display}](#{anchor})") + return "\n".join(toc_lines) def main(): @@ -565,17 +534,23 @@ def main(): # --- OpenRouter --- model_list_markdown += "### [OpenRouter](https://openrouter.ai)\n\n" if openrouter_models: - provider_limits = get_human_limits(openrouter_models[0]) # Limits are the same for all free models here + provider_limits = get_human_limits( + openrouter_models[0] + ) + model_list_markdown += "**Limits:**\n\n" model_list_markdown += f"[{provider_limits}
1000 requests/day with $10 credit balance](https://openrouter.ai/docs/api-reference/limits)\n\n" model_list_markdown += "Models share a common quota.\n\n" for model in openrouter_models: - model_list_markdown += f"- [{model['name']}](https://openrouter.ai/{model['id']})\n" + model_list_markdown += ( + f"- [{model['name']}](https://openrouter.ai/{model['id']})\n" + ) model_list_markdown += "\n" - # --- Google AI Studio --- model_list_markdown += "### [Google AI Studio](https://aistudio.google.com)\n\n" - model_list_markdown += "Data is used for training when used outside of the UK/CH/EEA/EU.\n\n" + model_list_markdown += ( + "Data is used for training when used outside of the UK/CH/EEA/EU.\n\n" + ) model_list_markdown += "\n" gemini_text_models = [ @@ -661,7 +636,9 @@ def main(): # Write text models to table for model in gemini_text_models: limits_str = get_human_limits(model) - model_list_markdown += f"\n" + model_list_markdown += ( + f"\n" + ) # Write embedding models to table first_embedding = True @@ -669,34 +646,48 @@ def main(): limits_str = get_human_limits(model) model_list_markdown += f"" if first_embedding: - model_list_markdown += f'' - first_embedding = False + model_list_markdown += f'' + first_embedding = False model_list_markdown += "\n" model_list_markdown += "
Model NameModel Limits
{model['name']}{limits_str}
{model['name']}{limits_str}
{model['name']}{limits_str}
100 content/batch
Shared Quota		{limits_str}
100 content/batch
Shared Quota
\n\n" # --- NVIDIA NIM --- - model_list_markdown += "### [NVIDIA NIM](https://build.nvidia.com/explore/discover)\n\n" - model_list_markdown += "Phone number verification required. Models tend to be context window limited.\n\n" - model_list_markdown += "- [Various open models](https://build.nvidia.com/models) (40 requests/minute)\n" + model_list_markdown += ( + "### [NVIDIA NIM](https://build.nvidia.com/explore/discover)\n\n" + ) + model_list_markdown += "Phone number verification required.\n" + model_list_markdown += "Models tend to be context window limited.\n\n" + model_list_markdown += "**Limits:** 40 requests/minute\n\n" + model_list_markdown += "- [Various open models](https://build.nvidia.com/models)\n" model_list_markdown += "\n" # --- Mistral (La Plateforme) --- - model_list_markdown += "### [Mistral (La Plateforme)](https://console.mistral.ai/)\n\n" - model_list_markdown += "Free tier (Experiment plan) requires opting into data training, requires phone number verification.\n\n" - model_list_markdown += "- [Open and Proprietary Mistral models](https://docs.mistral.ai/getting-started/models/models_overview/) (1 request/second, 500,000 tokens/minute, 1,000,000,000 tokens/month)\n" + model_list_markdown += ( + "### [Mistral (La Plateforme)](https://console.mistral.ai/)\n\n" + ) + model_list_markdown += "* Free tier (Experiment plan) requires opting into data training\n" + model_list_markdown += "* Requires phone number verification.\n\n" + model_list_markdown += "**Limits (per-model):** 1 request/second, 500,000 tokens/minute, 1,000,000,000 tokens/month\n\n" + model_list_markdown += "- [Open and Proprietary Mistral models](https://docs.mistral.ai/getting-started/models/models_overview/)\n" model_list_markdown += "\n" # --- Mistral (Codestral) --- - model_list_markdown += "### [Mistral (Codestral)](https://codestral.mistral.ai/)\n\n" - model_list_markdown += "Currently free to use, monthly subscription based, requires phone number verification.\n\n" - model_list_markdown += "- Codestral (30 requests/minute, 2,000 requests/day)\n" + model_list_markdown += ( + "### [Mistral (Codestral)](https://codestral.mistral.ai/)\n\n" + ) + model_list_markdown += "* Currently free to use\n" + model_list_markdown += "* Monthly subscription based\n" + model_list_markdown += "* Requires phone number verification\n\n" + model_list_markdown += "**Limits:** 30 requests/minute, 2,000 requests/day\n\n" + model_list_markdown += "- Codestral\n" model_list_markdown += "\n" # --- HuggingFace Serverless Inference --- - model_list_markdown += "### [HuggingFace Serverless Inference](https://huggingface.co/docs/api-inference/en/index)\n\n" - model_list_markdown += "Limited to models smaller than 10GB. Some popular models are supported even if they exceed 10GB.\n\n" - model_list_markdown += "- Various open models ([Variable credits per month, currently $0.10](https://huggingface.co/docs/api-inference/pricing))\n" + model_list_markdown += "### [HuggingFace Inference Providers](https://huggingface.co/docs/inference-providers/en/index)\n\n" + model_list_markdown += "HuggingFace Serverless Inference limited to models smaller than 10GB. Some popular models are supported even if they exceed 10GB.\n\n" + model_list_markdown += "**Limits:** [$0.10/month in credits](https://huggingface.co/docs/inference-providers/en/pricing)\n\n" + model_list_markdown += "- Various open models across supported providers\n" model_list_markdown += "\n" # --- Cerebras --- @@ -710,7 +701,9 @@ def main(): {"name": "Llama 3.3 70B", "limits_text": cerebras_limit_text}, ] for model in cerebras_models: - model_list_markdown += f"{model['name']}{model['limits_text']}\n" + model_list_markdown += ( + f"{model['name']}{model['limits_text']}\n" + ) model_list_markdown += "\n\n" # --- Groq --- @@ -719,17 +712,23 @@ def main(): model_list_markdown += "\n" for model in groq_models: limits_str = get_human_limits(model) - model_list_markdown += f"\n" + model_list_markdown += ( + f"\n" + ) model_list_markdown += "
Model NameModel Limits
{model['name']}{limits_str}
{model['name']}{limits_str}
\n" model_list_markdown += "\n" # --- OVH AI Endpoints --- - model_list_markdown += "### [OVH AI Endpoints (Free Beta)](https://endpoints.ai.cloud.ovh.net/)\n\n" + model_list_markdown += ( + "### [OVH AI Endpoints (Free Beta)](https://endpoints.ai.cloud.ovh.net/)\n\n" + ) if ovh_models: model_list_markdown += "\n" for model in ovh_models: limits_str = get_human_limits(model) - model_list_markdown += f"\n" + model_list_markdown += ( + f"\n" + ) model_list_markdown += "
Model NameModel Limits
{model['name']}{limits_str}
{model['name']}{limits_str}
\n" model_list_markdown += "\n" @@ -751,14 +750,15 @@ def main(): "urlId": "deepseek-r1-distilled-llama-70b-free", }, ] - model_list_markdown += "### [Together](https://together.ai)\n\n" - model_list_markdown += "Up to 60 requests/minute.\n\n" + model_list_markdown += "### [Together (Free)](https://together.ai)\n\n" + model_list_markdown += "**Limits:** Up to 60 requests/minute\n\n" if together_models: for model in together_models: - model_list_markdown += f"- [{model['name']}](https://together.ai/{model['urlId']})\n" + model_list_markdown += ( + f"- [{model['name']}](https://together.ai/{model['urlId']})\n" + ) model_list_markdown += "\n" - # --- Cohere --- cohere_models = [ {"id": "command-a-03-2025", "name": "Command-A"}, @@ -771,6 +771,7 @@ def main(): {"id": "c4ai-aya-vision-32b", "name": "Aya Vision 32B"}, ] model_list_markdown += "### [Cohere](https://cohere.com)\n\n" + model_list_markdown += "**Limits:**\n\n" model_list_markdown += "[20 requests/minute
1,000 requests/month](https://docs.cohere.com/docs/rate-limits)\n\n" model_list_markdown += "Models share a common quota.\n\n" if cohere_models: @@ -779,8 +780,11 @@ def main(): model_list_markdown += "\n" # --- GitHub Models --- - model_list_markdown += "### [GitHub Models](https://github.com/marketplace/models)\n\n" - model_list_markdown += "Extremely restrictive input/output token limits. [Rate limits dependent on Copilot subscription tier (Free/Pro/Business/Enterprise)](https://docs.github.com/en/github-models/prototyping-with-ai-models#rate-limits)\n\n" + model_list_markdown += ( + "### [GitHub Models](https://github.com/marketplace/models)\n\n" + ) + model_list_markdown += "Extremely restrictive input/output token limits.\n\n" + model_list_markdown += "**Limits:** [Dependent on Copilot subscription tier (Free/Pro/Pro+/Business/Enterprise)](https://docs.github.com/en/github-models/prototyping-with-ai-models#rate-limits)\n\n" if github_models: for model in github_models: model_list_markdown += f"- {model['name']}\n" @@ -788,21 +792,23 @@ def main(): # --- Chutes --- model_list_markdown += "### [Chutes](https://chutes.ai/)\n\n" - model_list_markdown += "Distributed, decentralized crypto-based compute. Data is sent to individual hosts.\n\n" + model_list_markdown += "Distributed, decentralized crypto-based compute.\n" + model_list_markdown += "Data is sent to individual hosts.\n\n" if chutes_models: for model in chutes_models: model_list_markdown += f"- {model['name']}\n" model_list_markdown += "\n" # --- Cloudflare Workers AI --- - model_list_markdown += "### [Cloudflare Workers AI](https://developers.cloudflare.com/workers-ai)\n\n" - model_list_markdown += "[10,000 neurons/day](https://developers.cloudflare.com/workers-ai/platform/pricing/#free-allocation)\n\n" + model_list_markdown += ( + "### [Cloudflare Workers AI](https://developers.cloudflare.com/workers-ai)\n\n" + ) + model_list_markdown += "**Limits:** [10,000 neurons/day](https://developers.cloudflare.com/workers-ai/platform/pricing/#free-allocation)\n\n" if cloudflare_models: for model in cloudflare_models: model_list_markdown += f"- {model['name']}\n" model_list_markdown += "\n" - # --- Google Cloud Vertex AI --- vertex_llama_models = [ { @@ -881,13 +887,12 @@ def main(): # Write Llama models to table if vertex_llama_models: - for model in vertex_llama_models: + for model in vertex_llama_models: limits_str = get_human_limits(model) model_list_markdown += f'{model['name']}{limits_str}
Free during preview\n' model_list_markdown += "\n\n" - # --- Trial Providers Section Generation --- trial_list_markdown = "" @@ -1006,26 +1011,26 @@ def main(): trial_list_markdown += f"- {model['name']}\n" trial_list_markdown += "\n" - if MISSING_MODELS: logger.warning("Missing models:") logger.warning( "\n" + "\n".join([f'"{model}": "{model}",' for model in MISSING_MODELS]) ) - script_dir = os.path.dirname(os.path.abspath(__file__)) with open(os.path.join(script_dir, "README_template.md"), "r") as f: readme = f.read() warning = """ """ + initial_templated = ( + (warning + readme) + .replace("{{MODEL_LIST}}", model_list_markdown) + .replace("{{TRIAL_LIST_MARKDOWN}}", trial_list_markdown) + ) + toc_markdown = generate_toc(initial_templated) with open(os.path.join(script_dir, "..", "README.md"), "w") as f: - f.write( - (warning + readme) - .replace("{{MODEL_LIST}}", model_list_markdown) - .replace("{{TRIAL_LIST_MARKDOWN}}", trial_list_markdown) - ) + f.write(initial_templated.replace("{{TOC}}", toc_markdown)) logger.info("Wrote models to README.md")