v27.4.0+incompatible

Compare Source

v0.19.0

Compare Source

What's Changed

• Update prompt in json_no_schema snippet by @​pcoet in https://github.com/google/generative-ai-go/pull/210
• Fix typo in resumable.go by @​windmaple in https://github.com/google/generative-ai-go/pull/215
• genai: fix text of error message by @​eliben in https://github.com/google/generative-ai-go/pull/216
• all: fix license headers and add a check by @​jba in https://github.com/google/generative-ai-go/pull/218
• genai: test invalid text file encoding by @​eliben in https://github.com/google/generative-ai-go/pull/224
• genai: omit empty text parts from session history by @​jba in https://github.com/google/generative-ai-go/pull/226
• genai: add error check to example by @​jba in https://github.com/google/generative-ai-go/pull/227
• genai: add file with 1251 encoding to test error by @​eliben in https://github.com/google/generative-ai-go/pull/228
• genai: fix link to MIME types doc by @​jba in https://github.com/google/generative-ai-go/pull/230
• genai: bump version for v0.19.0 release by @​eliben in https://github.com/google/generative-ai-go/pull/235

New Contributors

• @​pcoet made their first contribution in https://github.com/google/generative-ai-go/pull/210
• @​windmaple made their first contribution in https://github.com/google/generative-ai-go/pull/215

Full Changelog: google/generative-ai-go@v0.18.0...v0.19.0

v1.8.9

Compare Source

What's Changed

• Skip ignored private fields by @​rafaelhl in https://github.com/magiconair/properties/pull/61

New Contributors

• @​rafaelhl made their first contribution in https://github.com/magiconair/properties/pull/61

Full Changelog: magiconair/properties@v1.8.8...v1.8.9

v1.8.8

Compare Source

What's Changed

• migrate from deprecated ioutil dependency by @​mbenson in https://github.com/magiconair/properties/pull/73
• Load from io.Reader by @​mbenson in https://github.com/magiconair/properties/pull/74

New Contributors

• @​mbenson made their first contribution in https://github.com/magiconair/properties/pull/73

Full Changelog: magiconair/properties@v1.8.7...v1.8.8

v0.5.1

Compare Source

What's Changed

• Fixed issue where Ollama's API would generate JSON output when specifying "format": null
• Fixed issue where passing --format json to ollama run would cause an error

Full Changelog: ollama/ollama@v0.5.0...v0.5.1

v0.5.0

Compare Source

New models

• Llama 3.3: a new state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model.
• Snowflake Arctic Embed 2: Snowflake's frontier embedding model. Arctic Embed 2.0 adds multilingual support without sacrificing English performance or scalability.

Structured outputs

Ollama now supports structured outputs, making it possible to constrain a model's output to a specific format defined by a JSON schema. The Ollama Python and JavaScript libraries have been updated to support structured outputs, together with Ollama's OpenAI-compatible API endpoints.

REST API

To use structured outputs in Ollama's generate or chat APIs, provide a JSON schema object in the format parameter:

curl -X POST http://localhost:11434/api/chat -H "Content-Type: application/json" -d '{
  "model": "llama3.1",
  "messages": [{"role": "user", "content": "Tell me about Canada."}],
  "stream": false,
  "format": {
    "type": "object",
    "properties": {
      "name": {
        "type": "string"
      },
      "capital": {
        "type": "string"
      },
      "languages": {
        "type": "array",
        "items": {
          "type": "string"
        }
      }
    },
    "required": [
      "name",
      "capital", 
      "languages"
    ]
  }
}'

Python library

Using the Ollama Python library, pass in the schema as a JSON object to the format parameter as either dict or use Pydantic (recommended) to serialize the schema using model_json_schema().

from ollama import chat
from pydantic import BaseModel

class Country(BaseModel):
  name: str
  capital: str
  languages: list[str]

response = chat(
  messages=[
    {
      'role': 'user',
      'content': 'Tell me about Canada.',
    }
  ],
  model='llama3.1',
  format=Country.model_json_schema(),
)

country = Country.model_validate_json(response.message.content)
print(country)

JavaScript library

Using the Ollama JavaScript library, pass in the schema as a JSON object to the format parameter as either object or use Zod (recommended) to serialize the schema using zodToJsonSchema():

import ollama from 'ollama';
import { z } from 'zod';
import { zodToJsonSchema } from 'zod-to-json-schema';

const Country = z.object({
    name: z.string(),
    capital: z.string(), 
    languages: z.array(z.string()),
});

const response = await ollama.chat({
    model: 'llama3.1',
    messages: [{ role: 'user', content: 'Tell me about Canada.' }],
    format: zodToJsonSchema(Country),
});

const country = Country.parse(JSON.parse(response.message.content));
console.log(country);

What's Changed

• Fixed error importing model vocabulary files
• Experimental: new flag to set KV cache quantization to 4-bit (q4_0), 8-bit (q8_0) or 16-bit (f16). This reduces VRAM requirements for longer context windows.
  • To enable for all models, use OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q4_0 ollama serve
  • Note: in the future flash attention will be enabled by default where available, with kv cache quantization available on a per-model basis
  • Thank you @​sammcj for the contribution in in https://github.com/ollama/ollama/pull/7926

New Contributors

• @​dmayboroda made their first contribution in https://github.com/ollama/ollama/pull/7906
• @​Geometrein made their first contribution in https://github.com/ollama/ollama/pull/7908
• @​owboson made their first contribution in https://github.com/ollama/ollama/pull/7693

Full Changelog: ollama/ollama@v0.4.7...v0.5.0

v0.210.0

Compare Source

Features

• all: Auto-regenerate discovery clients (#​2886) (f446504)
• all: Auto-regenerate discovery clients (#​2888) (a990614)
• all: Auto-regenerate discovery clients (#​2892) (22e0d11)
• all: Auto-regenerate discovery clients (#​2893) (2657039)
• all: Auto-regenerate discovery clients (#​2895) (46fab24)
• all: Auto-regenerate discovery clients (#​2896) (efded92)
• Async refresh dry run in parallel with sync refresh (#​2849) (acd2581)
• option/internaloption: Add AuthCreds helper (#​2891) (6ee8349)

Bug Fixes

• gensupport: Context cancel error check (#​2890) (f57c82d)

v1.68.1: Release 1.68.1

Compare Source

Bug Fixes

• credentials/alts: avoid SRV and TXT lookups for handshaker service to work around hangs caused by buggy versions of systemd-resolved. (#​7861)

Dependencies

• Relax minimum Go version requirement from go1.22.7 to go1.22. (#​7831)

v0.19.3

Compare Source

What's Changed

• 🐛 Refactor certificate watcher to use polling, instead of fsnotify by @​k8s-infra-cherrypick-robot in https://github.com/kubernetes-sigs/controller-runtime/pull/3023
• 🐛 Use leader elector with client timeout by @​k8s-infra-cherrypick-robot in https://github.com/kubernetes-sigs/controller-runtime/pull/3031

Full Changelog: kubernetes-sigs/controller-runtime@v0.19.2...v0.19.3