From 76e050874701e0db4cf1ea930781302bf169c44b Mon Sep 17 00:00:00 2001 From: Hammad Shami Date: Thu, 23 Oct 2025 13:43:27 -0700 Subject: [PATCH] add helicone models + helicone model generation script --- package.json | 3 +- packages/core/script/generate-helicone.ts | 208 ++++++++++++++++++ providers/helicone/README.md | 20 ++ providers/helicone/logo.svg | 18 ++ .../helicone/models/chatgpt-4o-latest.toml | 22 ++ .../helicone/models/claude-3.5-haiku.toml | 23 ++ .../helicone/models/claude-3.5-sonnet-v2.toml | 23 ++ .../helicone/models/claude-3.7-sonnet.toml | 23 ++ .../helicone/models/claude-4.5-haiku.toml | 23 ++ .../helicone/models/claude-4.5-sonnet.toml | 23 ++ .../helicone/models/claude-opus-4-1.toml | 23 ++ providers/helicone/models/claude-opus-4.toml | 23 ++ .../helicone/models/claude-sonnet-4.toml | 23 ++ .../models/deepseek-r1-distill-llama-70b.toml | 21 ++ .../helicone/models/deepseek-reasoner.toml | 22 ++ .../models/deepseek-tng-r1t2-chimera.toml | 21 ++ .../models/deepseek-v3.1-terminus.toml | 22 ++ providers/helicone/models/deepseek-v3.2.toml | 21 ++ providers/helicone/models/deepseek-v3.toml | 22 ++ .../models/ernie-4.5-21b-a3b-thinking.toml | 21 ++ .../models/gemini-2.5-flash-lite.toml | 23 ++ .../helicone/models/gemini-2.5-flash.toml | 23 ++ providers/helicone/models/gemini-2.5-pro.toml | 23 ++ providers/helicone/models/gemma-3-12b-it.toml | 21 ++ providers/helicone/models/gemma2-9b-it.toml | 21 ++ providers/helicone/models/glm-4.6.toml | 21 ++ providers/helicone/models/gpt-4.1-mini.toml | 22 ++ providers/helicone/models/gpt-4.1-nano.toml | 22 ++ providers/helicone/models/gpt-4.1.toml | 22 ++ providers/helicone/models/gpt-4o-mini.toml | 22 ++ providers/helicone/models/gpt-4o.toml | 22 ++ .../helicone/models/gpt-5-chat-latest.toml | 22 ++ providers/helicone/models/gpt-5-mini.toml | 22 ++ providers/helicone/models/gpt-5-nano.toml | 22 ++ providers/helicone/models/gpt-5.toml | 22 ++ providers/helicone/models/gpt-oss-120b.toml | 21 ++ providers/helicone/models/gpt-oss-20b.toml | 21 ++ providers/helicone/models/grok-3-mini.toml | 22 ++ providers/helicone/models/grok-3.toml | 22 ++ .../models/grok-4-fast-non-reasoning.toml | 22 ++ .../models/grok-4-fast-reasoning.toml | 22 ++ providers/helicone/models/grok-4.toml | 22 ++ .../helicone/models/grok-code-fast-1.toml | 22 ++ .../models/hermes-2-pro-llama-3-8b.toml | 21 ++ providers/helicone/models/kimi-k2-0905.toml | 22 ++ .../helicone/models/kimi-k2-instruct.toml | 21 ++ providers/helicone/models/kimi-k2.toml | 22 ++ .../helicone/models/llama-3.1-8b-instant.toml | 21 ++ .../models/llama-3.1-8b-instruct-turbo.toml | 21 ++ .../models/llama-3.1-8b-instruct.toml | 21 ++ .../models/llama-3.3-70b-instruct.toml | 21 ++ .../models/llama-3.3-70b-versatile.toml | 21 ++ .../helicone/models/llama-4-maverick.toml | 21 ++ providers/helicone/models/llama-4-scout.toml | 21 ++ providers/helicone/models/llama-guard-4.toml | 21 ++ .../models/llama-prompt-guard-2-22m.toml | 21 ++ .../models/llama-prompt-guard-2-86m.toml | 21 ++ providers/helicone/models/mistral-nemo.toml | 21 ++ providers/helicone/models/mistral-small.toml | 21 ++ providers/helicone/models/o3-mini.toml | 22 ++ providers/helicone/models/o3-pro.toml | 21 ++ providers/helicone/models/o3.toml | 22 ++ providers/helicone/models/o4-mini.toml | 22 ++ .../models/qwen2.5-coder-7b-fast.toml | 21 ++ .../models/qwen3-235b-a22b-thinking.toml | 21 ++ providers/helicone/models/qwen3-30b-a3b.toml | 21 ++ providers/helicone/models/qwen3-32b.toml | 21 ++ .../models/qwen3-coder-30b-a3b-instruct.toml | 21 ++ providers/helicone/models/qwen3-coder.toml | 21 ++ .../models/qwen3-next-80b-a3b-instruct.toml | 21 ++ .../models/qwen3-vl-235b-a22b-instruct.toml | 21 ++ providers/helicone/provider.toml | 5 + 72 files changed, 1706 insertions(+), 1 deletion(-) create mode 100644 packages/core/script/generate-helicone.ts create mode 100644 providers/helicone/README.md create mode 100644 providers/helicone/logo.svg create mode 100644 providers/helicone/models/chatgpt-4o-latest.toml create mode 100644 providers/helicone/models/claude-3.5-haiku.toml create mode 100644 providers/helicone/models/claude-3.5-sonnet-v2.toml create mode 100644 providers/helicone/models/claude-3.7-sonnet.toml create mode 100644 providers/helicone/models/claude-4.5-haiku.toml create mode 100644 providers/helicone/models/claude-4.5-sonnet.toml create mode 100644 providers/helicone/models/claude-opus-4-1.toml create mode 100644 providers/helicone/models/claude-opus-4.toml create mode 100644 providers/helicone/models/claude-sonnet-4.toml create mode 100644 providers/helicone/models/deepseek-r1-distill-llama-70b.toml create mode 100644 providers/helicone/models/deepseek-reasoner.toml create mode 100644 providers/helicone/models/deepseek-tng-r1t2-chimera.toml create mode 100644 providers/helicone/models/deepseek-v3.1-terminus.toml create mode 100644 providers/helicone/models/deepseek-v3.2.toml create mode 100644 providers/helicone/models/deepseek-v3.toml create mode 100644 providers/helicone/models/ernie-4.5-21b-a3b-thinking.toml create mode 100644 providers/helicone/models/gemini-2.5-flash-lite.toml create mode 100644 providers/helicone/models/gemini-2.5-flash.toml create mode 100644 providers/helicone/models/gemini-2.5-pro.toml create mode 100644 providers/helicone/models/gemma-3-12b-it.toml create mode 100644 providers/helicone/models/gemma2-9b-it.toml create mode 100644 providers/helicone/models/glm-4.6.toml create mode 100644 providers/helicone/models/gpt-4.1-mini.toml create mode 100644 providers/helicone/models/gpt-4.1-nano.toml create mode 100644 providers/helicone/models/gpt-4.1.toml create mode 100644 providers/helicone/models/gpt-4o-mini.toml create mode 100644 providers/helicone/models/gpt-4o.toml create mode 100644 providers/helicone/models/gpt-5-chat-latest.toml create mode 100644 providers/helicone/models/gpt-5-mini.toml create mode 100644 providers/helicone/models/gpt-5-nano.toml create mode 100644 providers/helicone/models/gpt-5.toml create mode 100644 providers/helicone/models/gpt-oss-120b.toml create mode 100644 providers/helicone/models/gpt-oss-20b.toml create mode 100644 providers/helicone/models/grok-3-mini.toml create mode 100644 providers/helicone/models/grok-3.toml create mode 100644 providers/helicone/models/grok-4-fast-non-reasoning.toml create mode 100644 providers/helicone/models/grok-4-fast-reasoning.toml create mode 100644 providers/helicone/models/grok-4.toml create mode 100644 providers/helicone/models/grok-code-fast-1.toml create mode 100644 providers/helicone/models/hermes-2-pro-llama-3-8b.toml create mode 100644 providers/helicone/models/kimi-k2-0905.toml create mode 100644 providers/helicone/models/kimi-k2-instruct.toml create mode 100644 providers/helicone/models/kimi-k2.toml create mode 100644 providers/helicone/models/llama-3.1-8b-instant.toml create mode 100644 providers/helicone/models/llama-3.1-8b-instruct-turbo.toml create mode 100644 providers/helicone/models/llama-3.1-8b-instruct.toml create mode 100644 providers/helicone/models/llama-3.3-70b-instruct.toml create mode 100644 providers/helicone/models/llama-3.3-70b-versatile.toml create mode 100644 providers/helicone/models/llama-4-maverick.toml create mode 100644 providers/helicone/models/llama-4-scout.toml create mode 100644 providers/helicone/models/llama-guard-4.toml create mode 100644 providers/helicone/models/llama-prompt-guard-2-22m.toml create mode 100644 providers/helicone/models/llama-prompt-guard-2-86m.toml create mode 100644 providers/helicone/models/mistral-nemo.toml create mode 100644 providers/helicone/models/mistral-small.toml create mode 100644 providers/helicone/models/o3-mini.toml create mode 100644 providers/helicone/models/o3-pro.toml create mode 100644 providers/helicone/models/o3.toml create mode 100644 providers/helicone/models/o4-mini.toml create mode 100644 providers/helicone/models/qwen2.5-coder-7b-fast.toml create mode 100644 providers/helicone/models/qwen3-235b-a22b-thinking.toml create mode 100644 providers/helicone/models/qwen3-30b-a3b.toml create mode 100644 providers/helicone/models/qwen3-32b.toml create mode 100644 providers/helicone/models/qwen3-coder-30b-a3b-instruct.toml create mode 100644 providers/helicone/models/qwen3-coder.toml create mode 100644 providers/helicone/models/qwen3-next-80b-a3b-instruct.toml create mode 100644 providers/helicone/models/qwen3-vl-235b-a22b-instruct.toml create mode 100644 providers/helicone/provider.toml diff --git a/package.json b/package.json index eee04663..e50fde9d 100644 --- a/package.json +++ b/package.json @@ -15,7 +15,8 @@ } }, "scripts": { - "validate": "bun ./packages/core/script/validate.ts" + "validate": "bun ./packages/core/script/validate.ts", + "helicone:generate": "bun ./packages/core/script/generate-helicone.ts" }, "dependencies": { "@cloudflare/workers-types": "^4.20250801.0", diff --git a/packages/core/script/generate-helicone.ts b/packages/core/script/generate-helicone.ts new file mode 100644 index 00000000..9bc2f827 --- /dev/null +++ b/packages/core/script/generate-helicone.ts @@ -0,0 +1,208 @@ +#!/usr/bin/env bun + +import { z } from "zod"; +import path from "node:path"; +import { mkdir, rm, readdir, stat } from "node:fs/promises"; + +// Helicone public model registry endpoint +const DEFAULT_ENDPOINT = + "https://jawn.helicone.ai/v1/public/model-registry/models"; + +// Zod schemas to validate the Helicone response +const Pricing = z + .object({ + prompt: z.number().optional(), + completion: z.number().optional(), + cacheRead: z.number().optional(), + cacheWrite: z.number().optional(), + reasoning: z.number().optional(), + }) + .passthrough(); + +const Endpoint = z + .object({ + provider: z.string(), + providerSlug: z.string().optional(), + supportsPtb: z.boolean().optional(), + pricing: Pricing.optional(), + }) + .passthrough(); + +const ModelItem = z + .object({ + id: z.string(), + name: z.string(), + author: z.string().optional(), + contextLength: z.number().optional(), + maxOutput: z.number().optional(), + trainingDate: z.string().optional(), + description: z.string().optional(), + inputModalities: z.array(z.string()).optional(), + outputModalities: z.array(z.string()).optional(), + supportedParameters: z.array(z.string()).optional(), + endpoints: z.array(Endpoint).optional(), + }) + .passthrough(); + +const HeliconeResponse = z + .object({ + data: z.object({ + models: z.array(ModelItem), + total: z.number().optional(), + filters: z.any().optional(), + }), + }) + .passthrough(); + +function pickEndpoint(m: z.infer) { + if (!m.endpoints || m.endpoints.length === 0) return undefined; + // Prefer endpoint that matches author if available + if (m.author) { + const match = m.endpoints.find((e) => e.provider === m.author); + if (match) return match; + } + return m.endpoints[0]; +} + +function boolFromParams(params: string[] | undefined, keys: string[]): boolean { + if (!params) return false; + const set = new Set(params.map((p) => p.toLowerCase())); + return keys.some((k) => set.has(k.toLowerCase())); +} + +function sanitizeModalities(values: string[] | undefined): string[] { + if (!values) return ["text"]; // default to text + const allowed = new Set(["text", "audio", "image", "video", "pdf"]); + const out = values.map((v) => v.toLowerCase()).filter((v) => allowed.has(v)); + return out.length > 0 ? out : ["text"]; +} + +function formatToml(model: z.infer) { + const ep = pickEndpoint(model); + const pricing = ep?.pricing; + + const supported = model.supportedParameters ?? []; + + const nowISO = new Date().toISOString().slice(0, 10); + const rdRaw = model.trainingDate ? String(model.trainingDate) : nowISO; + const releaseDate = rdRaw.slice(0, 10); + const lastUpdated = releaseDate; + const knowledge = model.trainingDate + ? String(model.trainingDate).slice(0, 7) + : undefined; + + const attachment = false; // Not exposed by Helicone registry + const temperature = boolFromParams(supported, ["temperature"]); + const toolCall = boolFromParams(supported, ["tools", "tool_choice"]); + const reasoning = boolFromParams(supported, [ + "reasoning", + "include_reasoning", + ]); + + const inputMods = sanitizeModalities(model.inputModalities); + const outputMods = sanitizeModalities(model.outputModalities); + + const lines: string[] = []; + lines.push(`name = "${model.name.replaceAll('"', '\\"')}"`); + lines.push(`release_date = "${releaseDate}"`); + lines.push(`last_updated = "${lastUpdated}"`); + lines.push(`attachment = ${attachment}`); + lines.push(`reasoning = ${reasoning}`); + lines.push(`temperature = ${temperature}`); + lines.push(`tool_call = ${toolCall}`); + if (knowledge) lines.push(`knowledge = "${knowledge}"`); + lines.push(`open_weights = false`); + lines.push(""); + + if ( + pricing && + (pricing.prompt ?? + pricing.completion ?? + pricing.cacheRead ?? + pricing.cacheWrite ?? + (reasoning && pricing.reasoning)) !== undefined + ) { + lines.push(`[cost]`); + if (pricing.prompt !== undefined) lines.push(`input = ${pricing.prompt}`); + if (pricing.completion !== undefined) + lines.push(`output = ${pricing.completion}`); + if (reasoning && pricing.reasoning !== undefined) + lines.push(`reasoning = ${pricing.reasoning}`); + if (pricing.cacheRead !== undefined) + lines.push(`cache_read = ${pricing.cacheRead}`); + if (pricing.cacheWrite !== undefined) + lines.push(`cache_write = ${pricing.cacheWrite}`); + lines.push(""); + } + + const context = model.contextLength ?? 0; + const output = model.maxOutput ?? 4096; + lines.push(`[limit]`); + lines.push(`context = ${context}`); + lines.push(`output = ${output}`); + lines.push(""); + + lines.push(`[modalities]`); + lines.push(`input = [${inputMods.map((m) => `"${m}"`).join(", ")}]`); + lines.push(`output = [${outputMods.map((m) => `"${m}"`).join(", ")}]`); + + return lines.join("\n") + "\n"; +} + +async function main() { + const endpoint = DEFAULT_ENDPOINT; + + const outDir = path.join( + import.meta.dirname, + "..", + "..", + "..", + "providers", + "helicone", + "models", + ); + + const res = await fetch(endpoint); + if (!res.ok) { + console.error(`Failed to fetch registry: ${res.status} ${res.statusText}`); + process.exit(1); + } + const json = await res.json(); + + const parsed = HeliconeResponse.safeParse(json); + if (!parsed.success) { + parsed.error.cause = json; + console.error("Invalid Helicone response:", parsed.error.errors); + console.error("When parsing:", parsed.error.cause); + process.exit(1); + } + + const models = parsed.data.data.models; + + // Clean output directory: remove subfolders and existing TOML files + await mkdir(outDir, { recursive: true }); + for (const entry of await readdir(outDir)) { + const p = path.join(outDir, entry); + const st = await stat(p); + if (st.isDirectory()) { + await rm(p, { recursive: true, force: true }); + } else if (st.isFile() && entry.endsWith(".toml")) { + await rm(p, { force: true }); + } + } + let created = 0; + + for (const m of models) { + const fileSafeId = m.id.replaceAll("/", "-"); + const filePath = path.join(outDir, `${fileSafeId}.toml`); + const toml = formatToml(m); + await Bun.write(filePath, toml); + created++; + } + + console.log( + `Generated ${created} model file(s) under providers/helicone/models/*.toml`, + ); +} + +await main(); diff --git a/providers/helicone/README.md b/providers/helicone/README.md new file mode 100644 index 00000000..b9552a25 --- /dev/null +++ b/providers/helicone/README.md @@ -0,0 +1,20 @@ +Helicone Models + +Generate model TOMLs from Helicone’s public registry. + +Prerequisites +- Install Bun: https://bun.sh + +Commands +- Generate files: `bun run helicone:generate` +- Validate configs: `bun validate` + +Details +- Source endpoint: `https://jawn.helicone.ai/v1/public/model-registry/models` +- Output path: `providers/helicone/models/.toml` (flat, no provider folders) +- Dates: `release_date`/`last_updated` use `YYYY-MM-DD`; `knowledge` uses `YYYY-MM`. +- Pricing: writes `cost.reasoning` only when `reasoning = true`. +- Modalities: sanitized to `["text", "audio", "image", "video", "pdf"]`. + +Notes +- The generator cleans the output folder before writing: removes any nested provider folders and existing TOML files to keep Model IDs flat (e.g., `claude-3.5-haiku`). diff --git a/providers/helicone/logo.svg b/providers/helicone/logo.svg new file mode 100644 index 00000000..6ee7cb9c --- /dev/null +++ b/providers/helicone/logo.svg @@ -0,0 +1,18 @@ + + + + + + + + + + + + + + + + + + diff --git a/providers/helicone/models/chatgpt-4o-latest.toml b/providers/helicone/models/chatgpt-4o-latest.toml new file mode 100644 index 00000000..45fb2da6 --- /dev/null +++ b/providers/helicone/models/chatgpt-4o-latest.toml @@ -0,0 +1,22 @@ +name = "OpenAI ChatGPT-4o" +release_date = "2024-08-14" +last_updated = "2024-08-14" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-08" +open_weights = false + +[cost] +input = 5 +output = 20 +cache_read = 2.5 + +[limit] +context = 128000 +output = 16384 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-3.5-haiku.toml b/providers/helicone/models/claude-3.5-haiku.toml new file mode 100644 index 00000000..c75b7790 --- /dev/null +++ b/providers/helicone/models/claude-3.5-haiku.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude 3.5 Haiku" +release_date = "2024-10-22" +last_updated = "2024-10-22" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-10" +open_weights = false + +[cost] +input = 0.7999999999999999 +output = 4 +cache_read = 0.08 +cache_write = 1 + +[limit] +context = 200000 +output = 8192 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-3.5-sonnet-v2.toml b/providers/helicone/models/claude-3.5-sonnet-v2.toml new file mode 100644 index 00000000..bfb72520 --- /dev/null +++ b/providers/helicone/models/claude-3.5-sonnet-v2.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude 3.5 Sonnet v2" +release_date = "2024-10-22" +last_updated = "2024-10-22" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-10" +open_weights = false + +[cost] +input = 3 +output = 15 +cache_read = 0.30000000000000004 +cache_write = 3.75 + +[limit] +context = 200000 +output = 8192 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-3.7-sonnet.toml b/providers/helicone/models/claude-3.7-sonnet.toml new file mode 100644 index 00000000..e559ec4f --- /dev/null +++ b/providers/helicone/models/claude-3.7-sonnet.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude 3.7 Sonnet" +release_date = "2025-02-19" +last_updated = "2025-02-19" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-02" +open_weights = false + +[cost] +input = 3 +output = 15 +cache_read = 0.30000000000000004 +cache_write = 3.75 + +[limit] +context = 200000 +output = 64000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-4.5-haiku.toml b/providers/helicone/models/claude-4.5-haiku.toml new file mode 100644 index 00000000..08ec4695 --- /dev/null +++ b/providers/helicone/models/claude-4.5-haiku.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude 4.5 Haiku" +release_date = "2025-10-01" +last_updated = "2025-10-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-10" +open_weights = false + +[cost] +input = 1 +output = 5 +cache_read = 0.09999999999999999 +cache_write = 1.25 + +[limit] +context = 200000 +output = 8192 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-4.5-sonnet.toml b/providers/helicone/models/claude-4.5-sonnet.toml new file mode 100644 index 00000000..907dc6e9 --- /dev/null +++ b/providers/helicone/models/claude-4.5-sonnet.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude Sonnet 4.5" +release_date = "2025-09-29" +last_updated = "2025-09-29" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 3 +output = 15 +cache_read = 0.30000000000000004 +cache_write = 3.75 + +[limit] +context = 200000 +output = 64000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-opus-4-1.toml b/providers/helicone/models/claude-opus-4-1.toml new file mode 100644 index 00000000..c288e139 --- /dev/null +++ b/providers/helicone/models/claude-opus-4-1.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude Opus 4.1" +release_date = "2025-08-05" +last_updated = "2025-08-05" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-08" +open_weights = false + +[cost] +input = 15 +output = 75 +cache_read = 1.5 +cache_write = 18.75 + +[limit] +context = 200000 +output = 32000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-opus-4.toml b/providers/helicone/models/claude-opus-4.toml new file mode 100644 index 00000000..0bc044e1 --- /dev/null +++ b/providers/helicone/models/claude-opus-4.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude Opus 4" +release_date = "2025-05-14" +last_updated = "2025-05-14" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-05" +open_weights = false + +[cost] +input = 15 +output = 75 +cache_read = 1.5 +cache_write = 18.75 + +[limit] +context = 200000 +output = 32000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/claude-sonnet-4.toml b/providers/helicone/models/claude-sonnet-4.toml new file mode 100644 index 00000000..a137ae6b --- /dev/null +++ b/providers/helicone/models/claude-sonnet-4.toml @@ -0,0 +1,23 @@ +name = "Anthropic: Claude Sonnet 4" +release_date = "2025-05-14" +last_updated = "2025-05-14" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-05" +open_weights = false + +[cost] +input = 3 +output = 15 +cache_read = 0.30000000000000004 +cache_write = 3.75 + +[limit] +context = 200000 +output = 64000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/deepseek-r1-distill-llama-70b.toml b/providers/helicone/models/deepseek-r1-distill-llama-70b.toml new file mode 100644 index 00000000..af5e0d9b --- /dev/null +++ b/providers/helicone/models/deepseek-r1-distill-llama-70b.toml @@ -0,0 +1,21 @@ +name = "DeepSeek R1 Distill Llama 70B" +release_date = "2025-01-20" +last_updated = "2025-01-20" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.19999999999999998 +output = 0.6 + +[limit] +context = 128000 +output = 4096 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/deepseek-reasoner.toml b/providers/helicone/models/deepseek-reasoner.toml new file mode 100644 index 00000000..70d3513d --- /dev/null +++ b/providers/helicone/models/deepseek-reasoner.toml @@ -0,0 +1,22 @@ +name = "DeepSeek Reasoner" +release_date = "2025-01-20" +last_updated = "2025-01-20" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.56 +output = 1.68 +cache_read = 0.07 + +[limit] +context = 128000 +output = 64000 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/deepseek-tng-r1t2-chimera.toml b/providers/helicone/models/deepseek-tng-r1t2-chimera.toml new file mode 100644 index 00000000..1884935d --- /dev/null +++ b/providers/helicone/models/deepseek-tng-r1t2-chimera.toml @@ -0,0 +1,21 @@ +name = "DeepSeek TNG R1T2 Chimera" +release_date = "2025-07-02" +last_updated = "2025-07-02" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-07" +open_weights = false + +[cost] +input = 0.3 +output = 1.2 + +[limit] +context = 130000 +output = 163840 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/deepseek-v3.1-terminus.toml b/providers/helicone/models/deepseek-v3.1-terminus.toml new file mode 100644 index 00000000..288a0ebc --- /dev/null +++ b/providers/helicone/models/deepseek-v3.1-terminus.toml @@ -0,0 +1,22 @@ +name = "DeepSeek V3.1 Terminus" +release_date = "2025-09-22" +last_updated = "2025-09-22" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.27 +output = 1 +cache_read = 0.21600000000000003 + +[limit] +context = 128000 +output = 16384 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/deepseek-v3.2.toml b/providers/helicone/models/deepseek-v3.2.toml new file mode 100644 index 00000000..0a2d1d59 --- /dev/null +++ b/providers/helicone/models/deepseek-v3.2.toml @@ -0,0 +1,21 @@ +name = "DeepSeek V3.2" +release_date = "2025-09-22" +last_updated = "2025-09-22" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.27 +output = 0.41 + +[limit] +context = 163840 +output = 65536 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/deepseek-v3.toml b/providers/helicone/models/deepseek-v3.toml new file mode 100644 index 00000000..22628350 --- /dev/null +++ b/providers/helicone/models/deepseek-v3.toml @@ -0,0 +1,22 @@ +name = "DeepSeek V3" +release_date = "2024-12-26" +last_updated = "2024-12-26" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-12" +open_weights = false + +[cost] +input = 0.56 +output = 1.68 +cache_read = 0.07 + +[limit] +context = 128000 +output = 8192 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/ernie-4.5-21b-a3b-thinking.toml b/providers/helicone/models/ernie-4.5-21b-a3b-thinking.toml new file mode 100644 index 00000000..fdf30447 --- /dev/null +++ b/providers/helicone/models/ernie-4.5-21b-a3b-thinking.toml @@ -0,0 +1,21 @@ +name = "Baidu Ernie 4.5 21B A3B Thinking" +release_date = "2025-03-16" +last_updated = "2025-03-16" +attachment = false +reasoning = true +temperature = true +tool_call = false +knowledge = "2025-03" +open_weights = false + +[cost] +input = 0.07 +output = 0.28 + +[limit] +context = 128000 +output = 8000 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/gemini-2.5-flash-lite.toml b/providers/helicone/models/gemini-2.5-flash-lite.toml new file mode 100644 index 00000000..af209c73 --- /dev/null +++ b/providers/helicone/models/gemini-2.5-flash-lite.toml @@ -0,0 +1,23 @@ +name = "Google Gemini 2.5 Flash Lite" +release_date = "2025-07-22" +last_updated = "2025-07-22" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-07" +open_weights = false + +[cost] +input = 0.09999999999999999 +output = 0.39999999999999997 +cache_read = 0.024999999999999998 +cache_write = 0.09999999999999999 + +[limit] +context = 1048576 +output = 65535 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gemini-2.5-flash.toml b/providers/helicone/models/gemini-2.5-flash.toml new file mode 100644 index 00000000..2d060a0a --- /dev/null +++ b/providers/helicone/models/gemini-2.5-flash.toml @@ -0,0 +1,23 @@ +name = "Google Gemini 2.5 Flash" +release_date = "2025-06-17" +last_updated = "2025-06-17" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-06" +open_weights = false + +[cost] +input = 0.3 +output = 2.5 +cache_read = 0.075 +cache_write = 0.3 + +[limit] +context = 1048576 +output = 65535 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gemini-2.5-pro.toml b/providers/helicone/models/gemini-2.5-pro.toml new file mode 100644 index 00000000..8584c2e4 --- /dev/null +++ b/providers/helicone/models/gemini-2.5-pro.toml @@ -0,0 +1,23 @@ +name = "Google Gemini 2.5 Pro" +release_date = "2025-06-17" +last_updated = "2025-06-17" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-06" +open_weights = false + +[cost] +input = 1.25 +output = 10 +cache_read = 0.3125 +cache_write = 1.25 + +[limit] +context = 1048576 +output = 65536 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gemma-3-12b-it.toml b/providers/helicone/models/gemma-3-12b-it.toml new file mode 100644 index 00000000..ddfbdd55 --- /dev/null +++ b/providers/helicone/models/gemma-3-12b-it.toml @@ -0,0 +1,21 @@ +name = "Google Gemma 3 12B" +release_date = "2024-12-01" +last_updated = "2024-12-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-12" +open_weights = false + +[cost] +input = 0.049999999999999996 +output = 0.09999999999999999 + +[limit] +context = 131072 +output = 8192 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gemma2-9b-it.toml b/providers/helicone/models/gemma2-9b-it.toml new file mode 100644 index 00000000..34bfe5bd --- /dev/null +++ b/providers/helicone/models/gemma2-9b-it.toml @@ -0,0 +1,21 @@ +name = "Google Gemma 2" +release_date = "2024-06-25" +last_updated = "2024-06-25" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-06" +open_weights = false + +[cost] +input = 0.19999999999999998 +output = 0.19999999999999998 + +[limit] +context = 8192 +output = 8192 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/glm-4.6.toml b/providers/helicone/models/glm-4.6.toml new file mode 100644 index 00000000..08e948e5 --- /dev/null +++ b/providers/helicone/models/glm-4.6.toml @@ -0,0 +1,21 @@ +name = "Zai GLM-4.6" +release_date = "2024-07-18" +last_updated = "2024-07-18" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2024-07" +open_weights = false + +[cost] +input = 0.6 +output = 2.2 + +[limit] +context = 204800 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/gpt-4.1-mini.toml b/providers/helicone/models/gpt-4.1-mini.toml new file mode 100644 index 00000000..179fe138 --- /dev/null +++ b/providers/helicone/models/gpt-4.1-mini.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-4.1 Mini" +release_date = "2025-04-14" +last_updated = "2025-04-14" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-04" +open_weights = false + +[cost] +input = 0.39999999999999997 +output = 1.5999999999999999 +cache_read = 0.09999999999999999 + +[limit] +context = 1047576 +output = 32768 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-4.1-nano.toml b/providers/helicone/models/gpt-4.1-nano.toml new file mode 100644 index 00000000..782e8d14 --- /dev/null +++ b/providers/helicone/models/gpt-4.1-nano.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-4.1 Nano" +release_date = "2025-04-14" +last_updated = "2025-04-14" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-04" +open_weights = false + +[cost] +input = 0.09999999999999999 +output = 0.39999999999999997 +cache_read = 0.024999999999999998 + +[limit] +context = 1047576 +output = 32768 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-4.1.toml b/providers/helicone/models/gpt-4.1.toml new file mode 100644 index 00000000..60bd74d4 --- /dev/null +++ b/providers/helicone/models/gpt-4.1.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-4.1" +release_date = "2025-04-14" +last_updated = "2025-04-14" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-04" +open_weights = false + +[cost] +input = 2 +output = 8 +cache_read = 0.5 + +[limit] +context = 1047576 +output = 32768 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-4o-mini.toml b/providers/helicone/models/gpt-4o-mini.toml new file mode 100644 index 00000000..08366248 --- /dev/null +++ b/providers/helicone/models/gpt-4o-mini.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-4o-mini" +release_date = "2024-07-18" +last_updated = "2024-07-18" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-07" +open_weights = false + +[cost] +input = 0.15 +output = 0.6 +cache_read = 0.075 + +[limit] +context = 128000 +output = 16384 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-4o.toml b/providers/helicone/models/gpt-4o.toml new file mode 100644 index 00000000..210cfd84 --- /dev/null +++ b/providers/helicone/models/gpt-4o.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-4o" +release_date = "2024-05-13" +last_updated = "2024-05-13" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-05" +open_weights = false + +[cost] +input = 2.5 +output = 10 +cache_read = 1.25 + +[limit] +context = 128000 +output = 16384 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-5-chat-latest.toml b/providers/helicone/models/gpt-5-chat-latest.toml new file mode 100644 index 00000000..38e5b1b4 --- /dev/null +++ b/providers/helicone/models/gpt-5-chat-latest.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-5 Chat Latest" +release_date = "2024-09-30" +last_updated = "2024-09-30" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-09" +open_weights = false + +[cost] +input = 1.25 +output = 10 +cache_read = 0.12500000000000003 + +[limit] +context = 128000 +output = 16384 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-5-mini.toml b/providers/helicone/models/gpt-5-mini.toml new file mode 100644 index 00000000..85a1785b --- /dev/null +++ b/providers/helicone/models/gpt-5-mini.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-5 Mini" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.25 +output = 2 +cache_read = 0.024999999999999998 + +[limit] +context = 400000 +output = 128000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-5-nano.toml b/providers/helicone/models/gpt-5-nano.toml new file mode 100644 index 00000000..b7de8e61 --- /dev/null +++ b/providers/helicone/models/gpt-5-nano.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-5 Nano" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.049999999999999996 +output = 0.39999999999999997 +cache_read = 0.005 + +[limit] +context = 400000 +output = 128000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-5.toml b/providers/helicone/models/gpt-5.toml new file mode 100644 index 00000000..4cd80577 --- /dev/null +++ b/providers/helicone/models/gpt-5.toml @@ -0,0 +1,22 @@ +name = "OpenAI GPT-5" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 1.25 +output = 10 +cache_read = 0.12500000000000003 + +[limit] +context = 400000 +output = 128000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/gpt-oss-120b.toml b/providers/helicone/models/gpt-oss-120b.toml new file mode 100644 index 00000000..aedda742 --- /dev/null +++ b/providers/helicone/models/gpt-oss-120b.toml @@ -0,0 +1,21 @@ +name = "OpenAI GPT-OSS 120b" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 0.04 +output = 0.16 + +[limit] +context = 131072 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/gpt-oss-20b.toml b/providers/helicone/models/gpt-oss-20b.toml new file mode 100644 index 00000000..6ce7ea5b --- /dev/null +++ b/providers/helicone/models/gpt-oss-20b.toml @@ -0,0 +1,21 @@ +name = "OpenAI GPT-OSS 20b" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 0.049999999999999996 +output = 0.19999999999999998 + +[limit] +context = 131072 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/grok-3-mini.toml b/providers/helicone/models/grok-3-mini.toml new file mode 100644 index 00000000..1aeeb5ca --- /dev/null +++ b/providers/helicone/models/grok-3-mini.toml @@ -0,0 +1,22 @@ +name = "xAI Grok 3 Mini" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 0.3 +output = 0.5 +cache_read = 0.075 + +[limit] +context = 131072 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/grok-3.toml b/providers/helicone/models/grok-3.toml new file mode 100644 index 00000000..4ef9a20c --- /dev/null +++ b/providers/helicone/models/grok-3.toml @@ -0,0 +1,22 @@ +name = "xAI Grok 3" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 3 +output = 15 +cache_read = 0.75 + +[limit] +context = 131072 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/grok-4-fast-non-reasoning.toml b/providers/helicone/models/grok-4-fast-non-reasoning.toml new file mode 100644 index 00000000..2e006fdc --- /dev/null +++ b/providers/helicone/models/grok-4-fast-non-reasoning.toml @@ -0,0 +1,22 @@ +name = "xAI Grok 4 Fast Non-Reasoning" +release_date = "2025-09-19" +last_updated = "2025-09-19" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.19999999999999998 +output = 0.5 +cache_read = 0.049999999999999996 + +[limit] +context = 2000000 +output = 2000000 + +[modalities] +input = ["text", "image", "audio"] +output = ["text"] diff --git a/providers/helicone/models/grok-4-fast-reasoning.toml b/providers/helicone/models/grok-4-fast-reasoning.toml new file mode 100644 index 00000000..e1cf8a9d --- /dev/null +++ b/providers/helicone/models/grok-4-fast-reasoning.toml @@ -0,0 +1,22 @@ +name = "xAI: Grok 4 Fast Reasoning" +release_date = "2025-09-01" +last_updated = "2025-09-01" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.19999999999999998 +output = 0.5 +cache_read = 0.049999999999999996 + +[limit] +context = 2000000 +output = 2000000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/grok-4.toml b/providers/helicone/models/grok-4.toml new file mode 100644 index 00000000..f040d6fe --- /dev/null +++ b/providers/helicone/models/grok-4.toml @@ -0,0 +1,22 @@ +name = "xAI Grok 4" +release_date = "2024-07-09" +last_updated = "2024-07-09" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-07" +open_weights = false + +[cost] +input = 3 +output = 15 +cache_read = 0.75 + +[limit] +context = 256000 +output = 256000 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/grok-code-fast-1.toml b/providers/helicone/models/grok-code-fast-1.toml new file mode 100644 index 00000000..ba96a0c0 --- /dev/null +++ b/providers/helicone/models/grok-code-fast-1.toml @@ -0,0 +1,22 @@ +name = "xAI Grok Code Fast 1" +release_date = "2024-08-25" +last_updated = "2024-08-25" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-08" +open_weights = false + +[cost] +input = 0.19999999999999998 +output = 1.5 +cache_read = 0.02 + +[limit] +context = 256000 +output = 10000 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/hermes-2-pro-llama-3-8b.toml b/providers/helicone/models/hermes-2-pro-llama-3-8b.toml new file mode 100644 index 00000000..ced1f445 --- /dev/null +++ b/providers/helicone/models/hermes-2-pro-llama-3-8b.toml @@ -0,0 +1,21 @@ +name = "Hermes 2 Pro Llama 3 8B" +release_date = "2024-05-27" +last_updated = "2024-05-27" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-05" +open_weights = false + +[cost] +input = 0.14 +output = 0.14 + +[limit] +context = 131072 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/kimi-k2-0905.toml b/providers/helicone/models/kimi-k2-0905.toml new file mode 100644 index 00000000..39949956 --- /dev/null +++ b/providers/helicone/models/kimi-k2-0905.toml @@ -0,0 +1,22 @@ +name = "Kimi K2 Instruct (09/05)" +release_date = "2025-09-05" +last_updated = "2025-09-05" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.5 +output = 2 +cache_read = 0.39999999999999997 + +[limit] +context = 262144 +output = 16384 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/kimi-k2-instruct.toml b/providers/helicone/models/kimi-k2-instruct.toml new file mode 100644 index 00000000..7eff7e60 --- /dev/null +++ b/providers/helicone/models/kimi-k2-instruct.toml @@ -0,0 +1,21 @@ +name = "Kimi K2 Instruct" +release_date = "2025-09-28" +last_updated = "2025-09-28" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.5700000000000001 +output = 2.3 + +[limit] +context = 131072 +output = 131072 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/kimi-k2.toml b/providers/helicone/models/kimi-k2.toml new file mode 100644 index 00000000..380f3666 --- /dev/null +++ b/providers/helicone/models/kimi-k2.toml @@ -0,0 +1,22 @@ +name = "Kimi K2 Instruct" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 1 +output = 3 +cache_read = 0.5 + +[limit] +context = 131072 +output = 16384 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-3.1-8b-instant.toml b/providers/helicone/models/llama-3.1-8b-instant.toml new file mode 100644 index 00000000..1bbe01b2 --- /dev/null +++ b/providers/helicone/models/llama-3.1-8b-instant.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 3.1 8B Instant" +release_date = "2024-07-01" +last_updated = "2024-07-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-07" +open_weights = false + +[cost] +input = 0.049999999999999996 +output = 0.08 + +[limit] +context = 131072 +output = 32678 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-3.1-8b-instruct-turbo.toml b/providers/helicone/models/llama-3.1-8b-instruct-turbo.toml new file mode 100644 index 00000000..7ccbe50d --- /dev/null +++ b/providers/helicone/models/llama-3.1-8b-instruct-turbo.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 3.1 8B Instruct Turbo" +release_date = "2024-07-23" +last_updated = "2024-07-23" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-07" +open_weights = false + +[cost] +input = 0.02 +output = 0.03 + +[limit] +context = 128000 +output = 128000 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-3.1-8b-instruct.toml b/providers/helicone/models/llama-3.1-8b-instruct.toml new file mode 100644 index 00000000..7eec4f82 --- /dev/null +++ b/providers/helicone/models/llama-3.1-8b-instruct.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 3.1 8B Instruct" +release_date = "2024-07-23" +last_updated = "2024-07-23" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-07" +open_weights = false + +[cost] +input = 0.02 +output = 0.049999999999999996 + +[limit] +context = 16384 +output = 16384 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-3.3-70b-instruct.toml b/providers/helicone/models/llama-3.3-70b-instruct.toml new file mode 100644 index 00000000..81d4d605 --- /dev/null +++ b/providers/helicone/models/llama-3.3-70b-instruct.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 3.3 70B Instruct" +release_date = "2024-12-06" +last_updated = "2024-12-06" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-12" +open_weights = false + +[cost] +input = 0.13 +output = 0.39 + +[limit] +context = 128000 +output = 16400 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-3.3-70b-versatile.toml b/providers/helicone/models/llama-3.3-70b-versatile.toml new file mode 100644 index 00000000..6bb2caec --- /dev/null +++ b/providers/helicone/models/llama-3.3-70b-versatile.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 3.3 70B Versatile" +release_date = "2024-12-06" +last_updated = "2024-12-06" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2024-12" +open_weights = false + +[cost] +input = 0.59 +output = 0.7899999999999999 + +[limit] +context = 131072 +output = 32678 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-4-maverick.toml b/providers/helicone/models/llama-4-maverick.toml new file mode 100644 index 00000000..5e63cb47 --- /dev/null +++ b/providers/helicone/models/llama-4-maverick.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 4 Maverick 17B 128E" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.15 +output = 0.6 + +[limit] +context = 131072 +output = 8192 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/llama-4-scout.toml b/providers/helicone/models/llama-4-scout.toml new file mode 100644 index 00000000..b41392c0 --- /dev/null +++ b/providers/helicone/models/llama-4-scout.toml @@ -0,0 +1,21 @@ +name = "Meta Llama 4 Scout 17B 16E" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.08 +output = 0.3 + +[limit] +context = 131072 +output = 8192 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/llama-guard-4.toml b/providers/helicone/models/llama-guard-4.toml new file mode 100644 index 00000000..958a6da6 --- /dev/null +++ b/providers/helicone/models/llama-guard-4.toml @@ -0,0 +1,21 @@ +name = "Meta Llama Guard 4 12B" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.21 +output = 0.21 + +[limit] +context = 131072 +output = 1024 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/llama-prompt-guard-2-22m.toml b/providers/helicone/models/llama-prompt-guard-2-22m.toml new file mode 100644 index 00000000..9f0cae52 --- /dev/null +++ b/providers/helicone/models/llama-prompt-guard-2-22m.toml @@ -0,0 +1,21 @@ +name = "Meta Llama Prompt Guard 2 22M" +release_date = "2024-10-01" +last_updated = "2024-10-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-10" +open_weights = false + +[cost] +input = 0.01 +output = 0.01 + +[limit] +context = 512 +output = 2 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/llama-prompt-guard-2-86m.toml b/providers/helicone/models/llama-prompt-guard-2-86m.toml new file mode 100644 index 00000000..cd7f03bf --- /dev/null +++ b/providers/helicone/models/llama-prompt-guard-2-86m.toml @@ -0,0 +1,21 @@ +name = "Meta Llama Prompt Guard 2 86M" +release_date = "2024-10-01" +last_updated = "2024-10-01" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-10" +open_weights = false + +[cost] +input = 0.01 +output = 0.01 + +[limit] +context = 512 +output = 2 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/mistral-nemo.toml b/providers/helicone/models/mistral-nemo.toml new file mode 100644 index 00000000..928b7913 --- /dev/null +++ b/providers/helicone/models/mistral-nemo.toml @@ -0,0 +1,21 @@ +name = "Mistral Nemo" +release_date = "2024-07-18" +last_updated = "2024-07-18" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-07" +open_weights = false + +[cost] +input = 20 +output = 40 + +[limit] +context = 128000 +output = 16400 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/mistral-small.toml b/providers/helicone/models/mistral-small.toml new file mode 100644 index 00000000..264c8a5d --- /dev/null +++ b/providers/helicone/models/mistral-small.toml @@ -0,0 +1,21 @@ +name = "Mistral Small" +release_date = "2024-02-26" +last_updated = "2024-02-26" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-02" +open_weights = false + +[cost] +input = 75 +output = 200 + +[limit] +context = 128000 +output = 128000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/o3-mini.toml b/providers/helicone/models/o3-mini.toml new file mode 100644 index 00000000..02d7398a --- /dev/null +++ b/providers/helicone/models/o3-mini.toml @@ -0,0 +1,22 @@ +name = "OpenAI o3 Mini" +release_date = "2023-10-01" +last_updated = "2023-10-01" +attachment = false +reasoning = false +temperature = false +tool_call = true +knowledge = "2023-10" +open_weights = false + +[cost] +input = 1.1 +output = 4.4 +cache_read = 0.55 + +[limit] +context = 200000 +output = 100000 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/o3-pro.toml b/providers/helicone/models/o3-pro.toml new file mode 100644 index 00000000..252cd166 --- /dev/null +++ b/providers/helicone/models/o3-pro.toml @@ -0,0 +1,21 @@ +name = "OpenAI o3 Pro" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = false +temperature = false +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 20 +output = 80 + +[limit] +context = 200000 +output = 100000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/o3.toml b/providers/helicone/models/o3.toml new file mode 100644 index 00000000..9c41f53b --- /dev/null +++ b/providers/helicone/models/o3.toml @@ -0,0 +1,22 @@ +name = "OpenAI o3" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = false +temperature = false +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 2 +output = 8 +cache_read = 0.5 + +[limit] +context = 200000 +output = 100000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/o4-mini.toml b/providers/helicone/models/o4-mini.toml new file mode 100644 index 00000000..976e484c --- /dev/null +++ b/providers/helicone/models/o4-mini.toml @@ -0,0 +1,22 @@ +name = "OpenAI o4 Mini" +release_date = "2024-06-01" +last_updated = "2024-06-01" +attachment = false +reasoning = false +temperature = false +tool_call = true +knowledge = "2024-06" +open_weights = false + +[cost] +input = 1.1 +output = 4.4 +cache_read = 0.275 + +[limit] +context = 200000 +output = 100000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/qwen2.5-coder-7b-fast.toml b/providers/helicone/models/qwen2.5-coder-7b-fast.toml new file mode 100644 index 00000000..90b65e9b --- /dev/null +++ b/providers/helicone/models/qwen2.5-coder-7b-fast.toml @@ -0,0 +1,21 @@ +name = "Qwen2.5 Coder 7B fast" +release_date = "2024-09-15" +last_updated = "2024-09-15" +attachment = false +reasoning = false +temperature = true +tool_call = false +knowledge = "2024-09" +open_weights = false + +[cost] +input = 0.03 +output = 0.09 + +[limit] +context = 32000 +output = 8192 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-235b-a22b-thinking.toml b/providers/helicone/models/qwen3-235b-a22b-thinking.toml new file mode 100644 index 00000000..b1db0f58 --- /dev/null +++ b/providers/helicone/models/qwen3-235b-a22b-thinking.toml @@ -0,0 +1,21 @@ +name = "Qwen3 235B A22B Thinking" +release_date = "2025-07-25" +last_updated = "2025-07-25" +attachment = false +reasoning = true +temperature = true +tool_call = false +knowledge = "2025-07" +open_weights = false + +[cost] +input = 0.3 +output = 2.9000000000000004 + +[limit] +context = 262144 +output = 81920 + +[modalities] +input = ["text", "image", "video"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-30b-a3b.toml b/providers/helicone/models/qwen3-30b-a3b.toml new file mode 100644 index 00000000..03a6c19d --- /dev/null +++ b/providers/helicone/models/qwen3-30b-a3b.toml @@ -0,0 +1,21 @@ +name = "Qwen3 30B A3B" +release_date = "2025-06-01" +last_updated = "2025-06-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-06" +open_weights = false + +[cost] +input = 0.08 +output = 0.29 + +[limit] +context = 41000 +output = 41000 + +[modalities] +input = ["text", "image"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-32b.toml b/providers/helicone/models/qwen3-32b.toml new file mode 100644 index 00000000..3055f548 --- /dev/null +++ b/providers/helicone/models/qwen3-32b.toml @@ -0,0 +1,21 @@ +name = "Qwen3 32B" +release_date = "2025-04-28" +last_updated = "2025-04-28" +attachment = false +reasoning = true +temperature = true +tool_call = true +knowledge = "2025-04" +open_weights = false + +[cost] +input = 0.29 +output = 0.59 + +[limit] +context = 131072 +output = 40960 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-coder-30b-a3b-instruct.toml b/providers/helicone/models/qwen3-coder-30b-a3b-instruct.toml new file mode 100644 index 00000000..cec979ad --- /dev/null +++ b/providers/helicone/models/qwen3-coder-30b-a3b-instruct.toml @@ -0,0 +1,21 @@ +name = "Qwen3 Coder 30B A3B Instruct" +release_date = "2025-07-31" +last_updated = "2025-07-31" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-07" +open_weights = false + +[cost] +input = 0.09999999999999999 +output = 0.3 + +[limit] +context = 262144 +output = 262144 + +[modalities] +input = ["text"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-coder.toml b/providers/helicone/models/qwen3-coder.toml new file mode 100644 index 00000000..2a645bd5 --- /dev/null +++ b/providers/helicone/models/qwen3-coder.toml @@ -0,0 +1,21 @@ +name = "Qwen3 Coder 480B A35B Instruct Turbo" +release_date = "2025-07-23" +last_updated = "2025-07-23" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-07" +open_weights = false + +[cost] +input = 0.29 +output = 1.2 + +[limit] +context = 262144 +output = 16384 + +[modalities] +input = ["text", "image", "audio", "video"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-next-80b-a3b-instruct.toml b/providers/helicone/models/qwen3-next-80b-a3b-instruct.toml new file mode 100644 index 00000000..4ece05c7 --- /dev/null +++ b/providers/helicone/models/qwen3-next-80b-a3b-instruct.toml @@ -0,0 +1,21 @@ +name = "Qwen3 Next 80B A3B Instruct" +release_date = "2025-01-01" +last_updated = "2025-01-01" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-01" +open_weights = false + +[cost] +input = 0.14 +output = 1.4 + +[limit] +context = 262000 +output = 16384 + +[modalities] +input = ["text", "image", "video"] +output = ["text"] diff --git a/providers/helicone/models/qwen3-vl-235b-a22b-instruct.toml b/providers/helicone/models/qwen3-vl-235b-a22b-instruct.toml new file mode 100644 index 00000000..7ac6ba73 --- /dev/null +++ b/providers/helicone/models/qwen3-vl-235b-a22b-instruct.toml @@ -0,0 +1,21 @@ +name = "Qwen3 VL 235B A22B Instruct" +release_date = "2025-09-23" +last_updated = "2025-09-23" +attachment = false +reasoning = false +temperature = true +tool_call = true +knowledge = "2025-09" +open_weights = false + +[cost] +input = 0.3 +output = 1.5 + +[limit] +context = 256000 +output = 16384 + +[modalities] +input = ["text", "image", "video"] +output = ["text"] diff --git a/providers/helicone/provider.toml b/providers/helicone/provider.toml new file mode 100644 index 00000000..818fbfbf --- /dev/null +++ b/providers/helicone/provider.toml @@ -0,0 +1,5 @@ +name = "Helicone" +env = ["HELICONE_API_KEY"] +npm = "@ai-sdk/openai-compatible" +api = "https://ai-gateway.helicone.ai/v1" +doc = "https://helicone.ai/models"