Burada FreeLLMAPI proqramının 10 noyabr 2026 tarixindən bəri buraxılan 13 versiyası var; son versiya: 0.16.37. Hər versiya naşirin öz quraşdırma faylına aparır.
| Versiya | Tarix | Ölçü | Fayllar |
|---|---|---|---|
| 0.16.37Son versiya | 20 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriAdds cheapllm.shop as a provider: one free OpenAI-compatible model, DeepSeek V4.1 Flash 1017, with tools, JSON schema and image input. Add a cheapllm.shop key on the Keys page. Its model row arrives through the hosted catalog: Premium now, Free after the usual 30 days. Also: a provider that answers HTTP 200 with only an error body (such as a busy queue) now fails over with the right cooldown instead of counting as an empty answer. See #1863. |
| 0.16.36 | 19 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriAdds explicit OVH translation and summarization endpoints. Use GET /v1/text/models to see eligible specialists, POST /v1/text/translate to translate, or POST /v1/text/summarize to summarize. These routes stay out of general chat and automatic fallback. The existing Premium and 30-day Free catalog schedule is unchanged. See #2009. |
| 0.16.34 | 17 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriDesktop agents
Open Agents on macOS or Windows to see all 21 supported agents in one scrollable list. Search by name to narrow the list, and clear the search to see everyone again.
Choose Connect to FreeLLMAPI next to Claude Code, Cline, OpenCode, or another agent to set up its gateway connection. The app saves supported configurations and guides you through tools that require a manual step.
This update also fixes how the desktop app handles audio buffers for transcription uploads. |
| 0.16.33 | 17 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriAdds Meldport chat for GPT-6 Luna’s free alias. The adapter requests promotional funding only, checks the response’s funding and model identity, and supports streaming. Meldport GLM is not enabled.
Adds keyless OVHcloud Solaria-3 transcription for English PCM16 WAV uploads. Existing OVH Whisper transcription continues to use its original endpoint.
Model availability follows the existing signed catalog: Premium immediately on publication, Free after 30 days. No changes to existing model routing or catalog eligibility. |
| 0.16.32 | 17 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriSpeechFlow file transcription is now supported. In Keys, select SpeechFlow and enter both the key ID and the key secret. Free.ai native embeddings and two more SilvR Nemotron embedding sizes are also supported.
The five routes are delivered through the signed catalog; existing local disable preferences are respected. Catalog availability is unchanged: Premium gets new routes immediately and Free after 30 days. No change to Mistral or to existing provider limits. |
| 0.16.31 | 16 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriTested catalog media routes
- Add NovAI GLM 4.7 Flash chat support in both normal and streaming modes.
- Add ElectronHub FLUX.2 Klein Base 4B image generation.
- Add SilvR Whisper Large V3 transcription.
- Preserve strict model identity checks, bounded image decoding, shared provider allowances and rejection of unapproved routes.
These adapters support the separately published November 16 catalog update: ten tested additions and eight retired Nara routes disabled, net +2 enabled entries.
No catalog rows are hard-deleted. Mistral is unchanged. Premium receives new catalog entries immediately; Free retains the existing 30-day delay. Install this update to use the three new adapter-supported routes.
Existing upstream changes on main since v0.16.30 also include the Claude Desktop /v1 compatibility fix, Windows Store packaging, and documentation improvements. |
| 0.16.30 | 15 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriNative routing for the November 15 renewable-free catalog additions:
- SilvR AI: three embedding models, Whisper Large V3 Turbo transcription, and Gemini Decision 1 predicate questions.
- Cartesia Ink 2 transcription: PCM16 WAV upload over the native, proxy-aware WebSocket with finalization checks and no ffmpeg dependency.
- ElectronHub images: Qwen Image 2512 plus dispatch for the two existing catalog image routes; bounded image decoding and safe timeout/error handling.
Also includes the current main-branch fixes for Windows bootstrap root resolution, explicit non-streaming chat requests, key health evidence, the Codex model catalog endpoint, and advertising the router's effective context window in model listings. Concurrently merged README contributor credits are preserved.
Provider free allowances remain shared, renewable and subject to availability. No new app quotas, paid fallback, Mistral catalog changes or premium/Free release-timing changes. Catalog publication is separate: seven additions and one retired Kenari route disabled, with premium access immediately and Free eligibility after the existing 30-day window. |
| 0.16.29 | 14 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriNew providers
- Typecast: native SSFM 3.1, 3.0 and 2.1 text-to-speech.
- Cartesia: Sonic 3.6, 3.5 and 3 speech, plus Ink Whisper batch transcription.
- SilvR AI: OpenAI-compatible chat with normal and streaming model-identity checks.
The new providers appear in local key setup/import. Typecast Free output is non-commercial and requires attribution; Cartesia's commercial-use license is on a paid plan. Free allowances are shared account pools, not credits per model.
Native speech adapters preserve error/backoff information for failover, refuse credential redirects, bound downloads and validate actual audio / transcription responses. Chat cannot accidentally route to a speech model. This release also includes the recent URL-guard, Hugging Face video-download and request-lease fixes already merged into main.
Model rows are delivered separately through the signed catalog. Premium receives updates immediately; Free receives them after the existing 30-day aging rule. Mistral and local disabled choices are unchanged. |
| 0.16.28 | 13 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriNew catalog adapter support
- Cloudflare EmbeddingGemma 2: native text embeddings with fixed 768-dimensional vectors, batch ordering and model-identity validation. Uses its own embedding family so existing indexes never silently switch vector spaces.
- Google Gemini Decisions: native typed decisions using existing Google credentials, with validated answers, confidence and token usage.
- ElectronHub transcription: Voxtral Mini Transcribe 2 and Deepgram Nova 3 through the existing key pool and multipart speech endpoint.
Quota errors and retry hints remain available to the normal router. Existing providers, Mistral behavior and user overrides are unchanged.
The accompanying catalog update is delivered separately: ten additions, three retired routes disabled (net enabled +7), and corrected Speechify allowance metadata. Premium receives the Live catalog immediately; Free receives new entries through the existing 30-day release process.
Validation
Production builds, lint, migration round-trip tests, full server/CLI/client suites and desktop tests passed locally. Native adapters have regression coverage for malformed responses, model substitutions, vector dimensions and rate limits. Live catalog tests use the actual local application API without saving provider credentials. |
| 0.16.27 | 12 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriZro by MoonMath.ai
- Add your own Zro API key in the Keys dashboard; imports recognize ZRO_API_KEY and MOONMATH_API_KEY.
- OpenAI-compatible normal and streaming chat, tool calls and JSON output use the existing routing and failover behavior.
- Credential validation uses the authenticated model-list endpoint without spending inference tokens.
- Reject mismatched response model identities, credential redirects and in-band streaming errors.
- Account-wide Free usage resets daily; it is not a monthly cash grant. Model availability is managed through the signed catalog rather than bundled seeds.
Catalog compatibility
This release supplies the Zro adapter needed by the November 12 catalog update. The existing Premium-immediate / Free-after-30-days catalog logic is unchanged. Mistral and other providers' application behavior are unchanged.
The catalog publication is separate from the app release. The validated update holds three dated/non-renewing promotions rather than mislabeling them as recurring free, retains stored history, and adds only tested routes.
Local validation: full server/CLI/client suites, locale validation, migration round trips, lint, workspace build and desktop tests. Zro was also tested through the real app router against the Free account with an in-memory database. |
| 0.16.26 | 11 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriOutcrawl web search
- Add Outcrawl (api.outcrawl.ai) as a keyed web search engine (#1802). Its free plan gives 1,000 credits a month with no card. A search costs about 0.5 credit per result, so roughly 400 searches at the default 5 results. Key checks read the credit balance and spend nothing. News searches run as ordinary web searches. Thanks @heydryft.
- The engine arrives through the signed catalog: Premium gets it now, and Free gets it after the usual 30 days.
Routing
- Each chain can now have its own routing strategy (Manual order, Balanced, Smartest, Fastest, Most reliable) or inherit the global one (#1697, #1696). An auto:<axis> alias in the request still wins over the chain's setting. Thanks @navzo.
Fixes
- The monthly token budget bar keeps counting tokens after analytics retention deletes old request rows (#1782). History deleted before this update cannot be recovered. Thanks @kvianAR.
Maintenance
- Upgrade Zod to v4 and Vitest to v5 (#1779). Validation errors still return 400, now with Zod 4's wording. Thanks @thedevil11255.
- The CLI test suite passes on Windows (#1803). Thanks @leucds2018-hub.
Also in v0.16.25 (not listed in its notes)
- A model that answers "this model is overloaded / experiencing high demand" with a 5xx is skipped for the rest of the request instead of being retried on every key (#1781). Thanks @kvianAR.
- Gemini no longer rejects tool schemas that put items on a non-array schema (#1791, fixes #1790). Thanks @TanmayShah29.
- Settings → General has a "Start on system startup" switch on macOS and Windows (#1799). Thanks @MoriDanWork. |
| 0.16.25 | 11 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriFreeTTS: local personal BYOK speech
- Add FreeTTS key setup and import support for your own local account.
- Add native MP3 text-to-speech with voice selection, fixed-host credential-safe downloads, bounded responses, audio validation and rate-limit failover.
- Keep the provider's spoken watermark. FreeTTS Free is personal/non-commercial; local use does not waive its terms or permit reselling shared API access.
- The signed catalog adds one logical TTS entry, neural-tts. Voices are parameters, not thousands of separate models. Premium receives the catalog entry now; Free receives it after the existing 30-day aging rule. No gate changes or early model seeding.
- Keep every other provider/model row, Mistral settings, and existing user overrides unchanged.
Install this release to use the FreeTTS adapter. Add your own FreeTTS key locally; no provider key is bundled or published.
The release also includes the fixes already merged on main since v0.16.24, including Gemini schema compatibility, overloaded-model sibling-key fallback, and the Start on system startup toggle.
Linux installer builds are pinned to Ubuntu 24.04 for compatibility with the current packaging tools; AppImage, tar.xz, DEB and RPM remain available.
Provider details: https://freetts.org/developers · https://freetts.org/pricing · https://freetts.org/terms |
| 0.16.24 | 10 noyabr 2026 | — | 64 bit EXE (NSIS) Versiya qeydləriPowerTokens provider
- Add PowerTokens (api.powertokens.ai) as an OpenAI-compatible provider. Its glm-4.7-flash model is listed at $0 (0 Credits); chat, streaming and tool calling were tested through the app router.
- PowerTokens is often overloaded and answers 429 (code 1305). The router cools the key down and moves on, so treat it as an extra fallback rather than a primary route.
- Keys restricted to other models in the PowerTokens dashboard return 403 token_model_forbidden; allow glm-4.7-flash on the key.
The model arrives through the signed catalog. Premium receives the live catalog; Free receives the new row after the unchanged 30-day aging period. Thanks @PowerTokens for the original contribution (#1659).
Fixes
- Custom endpoints with several keys now rotate image, speech, transcription and embedding requests across those keys when one is disabled, out of budget, rate limited or rejected (#1738, fixes #1737). Thanks @w21180239.
- Free.ai is no longer offered for chat routing, and "Test" on a legacy non-chat model row explains that the platform does not support chat instead of reporting a missing key (#1769). Thanks @w21180239.
- "Probe now" gives reasoning models room for their hidden thinking (#1862).
- freellmapi setup-opencode stores the key in auth.json instead of an environment placeholder (#1861).
Docs
- Turkish README translation (#1805). Thanks @alkank. |