v0.8.2
Added
nudgessetting (default on). Set it tofalseand an unattendedoptiq coderun leaves stalls, repeated calls and overdue test runs to the model.- The
localtool takestimeout_seconds.
Changed
- An unattended
optiq coderun is no longer asked to review its work before it finishes. - The
localworker gets at most 40% of an unattended run's time, and never the last 10%. - A call repeated with nothing changed gets the "change approach" message at the third refusal, not the twentieth, and the run stops at ten.
Fixed
optiq code'srun_testsuses the repo's own runner (go test,cargo test,npm test) and counts Go results.- A timed-out command is stopped with everything it started.
- Any file the agent writes counts as a change.
run_testsfalls back topython -m unittestin a Python repo with no pytest installed.
v0.8.1
Added
optiq serveuses the tested speed kernels on Groot-19B-A3B-OptiQ-4bit.
Fixed
optiq convertquantizes the routed experts of an MoE model's MTP head when the source stores them fused. They stayed at bf16, so--mtprefused the head.- A model with no MTP head logs an info line, not an "injection failed" warning.
v0.8.0
One config, breaking. See Configuration.
Changed
- All settings live in one YAML file,
~/.optiq/config.yaml, or<repo>/.optiq/config.yamlfor a repo. The first run moves your old user files and keeps them as.bak, so you stay logged in. Move a repo's.optiq/code.jsonor.optiq/optiq.jsonyourself. optiq configlists everything.optiq code configis gone.- Every setting has one variable,
OPTIQ_<NAME>:OPTIQ_CODE_MODELis nowOPTIQ_MODEL, and so on. An old or unknownOPTIQ_variable is ignored with a warning. optiq convert:--target-bpwis--target-bits,--n-calibrationis--n-samples,--targetis gone.optiq eval:--reference-modeis--reference,--skip-klis gone.optiq prune-experts:--target-retentionis now--retainwith a fraction.--stream-expertstakesauto,onoroffeverywhere.optiq lab:--lab-portis--port.optiq code:-y's long form is--auto-approve.OPTIQ_LOWBIT_SEARCHisOPTIQ_LOW_BIT_SEARCH, andOPTIQ_SERVE_ON_GENERATION_DEATHisOPTIQ_ON_GENERATION_DEATH.- Codex launched through OptiQ reads its key from
OPTIQ_API_KEY, notOPTIQ_AUTH_TOKEN.
Removed
- 26 internal tuning settings, now fixed:
compact_headroom,compact_target,compact_floor_chars,compact_min_yield,compact_keep_recent,explore_limit,edit_fail_limit,dup_call_limit,max_truncations,mcp_defer,warm_cache,line_numbers,edit_tolerance,transport_retry_s,guidance_max_chars,local_tool_max_turns,local_tool_max_seconds,test_nudge_edits,flag_test_edits,server_ready_timeout,server_host,lowbit_search_max_bits,stream_scales_budget_gb,cluster_gpu_reserve_gb,cluster_cache_mb,fused_ce_budget_mb.
v0.7.11
Added
- OptiQ Code keeps a killed session's conversation, written every turn, so
-cresumes it. Each session also logs one line per model call: timing, tokens and status.
Fixed
optiq serve --mtpapplies the model's sampling penalties as plain serving does. It ignored them, so it could repeat itself and wrote different text. Structured output under--mtpuses plain decoding.optiq serve --mtpreportslengthwhen it runs out of tokens, and no longer goes one token overmax_tokens.
v0.7.10
Fixed
- The MTP head now loads on quants that store its experts one by one, such as Qwen3.5-35B-A3B and Ornith-1.5-35B-A3B. It was refused before, so they drafted without it.
run_testsin OptiQ Code retries its default aspython -m pytestwhen collection cannot import a module, so a repo with a module at its root no longer reports passing tests as failed.- OptiQ Code's final answer is written for the person reading it, rather than repeating its completion review item by item.
v0.7.9
Added
- Liquid's d1 decision models (
LiquidAI/d1-*and their quants):optiq.decision.D1, andoptiq serveanswers them on/v1/systemone. Yes/no, choice, score and estimate questions over text and JSON states; span reads and pictures not yet.
Fixed
- LFM3 models no longer abort the server on the first request handled off the loading thread ("There is no Stream(gpu, 0) in current thread").
optiq convertno longer fails with "load_tokenizer() got multiple values for argument 'tokenizer_config_extra'" when a model with custom code was loaded earlier in the same process.
Earlier releases
The 14 releases before v0.7.9. Click any version to expand it. Everything before v0.6.3 is in the archive.
v0.7.8
Added
- LFM3 (Liquid):
lfm3text models and the LFM3-VL vision front-end. Convert, eval,kv-cache,serveand images through the vision sidecar.
Fixed
optiq kv-cacheand the Python engine load vendored architectures (LFM3, Xing4.0, LLaDA2, Mistral 4); they failed with "Model type ... not supported".
v0.7.7
Added
- OptiQ Code sandbox, off by default:
/sandbox,--sandboxor"sandbox": truerunsbashandrun_testsunder sandbox-exec (macOS) or bwrap (Linux). Writes only to the repo, temp and package caches; credential files unreadable; secret environment variables removed; network only to an allowlist of package registries and git hosts, with a prompt for anything else.
Changed
- The Lab's
pythonandterminaltools share the same sandbox code: credential files are now unreadable there too, CPU, process and file-size limits apply, and Linux without a container runtime uses bwrap instead of running unconfined.
v0.7.6
Fixed
optiq run claudeno longer makes Claude Code warn that claude.ai connectors are disabled. They never load under an OptiQ token; locally configured MCP servers work as before.optiq eval --servedmatches in-process scoring again: it sends penalties as off instead of leaving them out, which let the server apply the model's recommended presence penalty.
v0.7.5
Changed
optiq servepasses a requested reasoning effort through to the model: any effort other thannone/minimal, or Anthropicthinking.type: enabled, turns thinking on, and the effort reaches templates that readreasoning_effort(Spark-X2.6, gpt-oss). A template that rejects the value (Spark-X2.6 takes onlylowandhigh) renders at its default instead of failing the request.
v0.7.4
Added
- OptiQ Code reads Agent Skills (
SKILL.mdfolders) from.agents/skillsand.claude/skills, in the project and your home directory. The agent loads one with the newskilltool when a task matches,/<name>runs one, and/skillslists them.
Fixed
- OptiQ Code counts tests run through
bash(for examplepython3 -m pytest) likerun_tests: the run reports their result instead of "0 passed / 0 failed", and a passing run is no longer nudged as stuck on the next turn. optiq servehonours the standard thinking controls:reasoning_effort(Chat Completions),reasoning.effort(Responses) andthinking.type(Messages)."none","minimal"or"disabled"turn thinking off; they were ignored, so the model reasoned anyway. An explicitchat_template_kwargsstill wins.
v0.7.3
Fixed
optiq run claude --cloudno longer makes Claude Code warn that its auto-compact window is too large: it is held to the 1M Claude Code accepts.- The Lab reads an OptiQ Cloud model's window from the model list. It asked about the local model's window instead, or got nothing on a cloud-only server, and ran with compaction off and an 8k reply budget.
v0.7.2
Changed
optiq run --cloudgives the agent the model's full listed context window (about 1M tokens) instead of capping it at 200,000; each agent decides when to compact.
Fixed
- Claude Code on an OptiQ Cloud model could print a reply starting with
{"type": "thinking", ...}: the Anthropic shim turned earlier thinking blocks into text. They now travel as reasoning, never as text.
v0.7.1
Changed
optiq serverefuses a--hostother than this machine unless it has an API key (--api-keyorOPTIQ_API_KEY), and then every request must carry that key asAuthorization: Bearerorx-api-key. On 127.0.0.1 nothing changes.optiq serve --cloud-onlyandoptiq cluster servefollow the same rule.optiq lab --hostexposes only the Lab UI; its model API stays on 127.0.0.1.
v0.7.0
Added
optiq servelists OptiQ Cloud's models asoptiq-cloud/<id>once signed in and answers them for every client.optiq serve --cloud-onlyserves only OptiQ Cloud models, with no local model loaded.optiq run <agent> --cloud <model>runs Claude Code, Codex, OpenCode, Mistral Vibe, Hermes Agent or OpenClaw on an OptiQ Cloud model.optiq cloud modelslists OptiQ Cloud's models with their prices and context windows.- OptiQ Code:
/boost <model>and theboost_modelsetting pick the cloud model Boost runs on;/boost offturns it off. - The Lab lists OptiQ Cloud's models, and
optiq labserves them when signed in with no local model.
Changed
- Claude Code and Codex receive OptiQ Cloud replies as a stream, including tool calls, instead of all at once.
Fixed
optiq run hermesfailed with Hermes Agent 0.16, which sends its key only to openai.com.
v0.6.7
Changed
- OptiQ Code Boost now runs through
optiq serveon the same Mac (the served modeloptiq-boost), not by calling OptiQ Cloud directly. - Boost needs
optiq serverunning and logged in withoptiq cloud login; without it, Code runs on the local model and says how to turn Boost on. optiq servestreams OptiQ Cloud's reply to OpenAI-format clients as it arrives, instead of buffering it.- Boost's local helper returns what it found: no cap on its answer, a time limit, and a final wrap-up turn.
- Boost passes the cloud model its own earlier reasoning on later turns.
Fixed
optiq serveputs back the reasoning that a client dropped from earlier tool-call turns, so the model sees its own reasoning and the prompt cache is reused on hybrid models.- Boost continues on the local model when OptiQ Cloud cannot answer, and retries a cut stream.
v0.6.6
Changed
- Lab hybrid search lists keyword matches first and gives the last places to meaning matches keyword search missed. Document search returns 10 results by default (was 5).
- Lab deep research follows the Thinking toggle and the thinking budget.
Fixed
- A Lab reply that ended inside the model's reasoning came back empty. The Lab now asks once for the answer.
- Rebuilding a meaning index at a new vector size mixed old and new vectors.
v0.6.5
Added
- Lab and OptiQ Code draft a bug report when they crash. Nothing is sent until you review it and choose Send.
- The Lab's feedback form asks what kind of feedback it is: something's wrong, an idea, or something else.
v0.6.4
Added
- Lab collections: upload documents and search them in chat and deep research, by keyword and by meaning.
optiq serve --embedding-modelserves/v1/embeddingsfrom a Qwen3-Embedding model next to the chat model.- MCP 2.x SDK support, alongside 1.x.
- OptiQ Code
temperatureandtop_psettings for your own endpoints.
Changed
- Sampling defaults come from each model's
generation_config, including presence and frequency penalties. OptiQ Code and the Lab no longer override them. optiq servechecks the port before loading the model.
Fixed
- Thinking budgets had no effect on mlx-lm 0.32. The Lab now sends one.
- The Lab read the context window from the wrong route and capped every turn at 8k tokens.
- Chunked verify never turned on in
optiq serve. - A failed server start could hang with the model loaded, or restart into the same busy port.
v0.6.3
Changed
optiq convertwrites the vision towers beside the model shards instead of underoptiq/, and copies the base model's image processor files. LM Studio downloads and oMLX now load the vision towers; the MTP head stays inoptiq/.
Fixed
- Chunked verify attention never ran: the GQA decode kernel installs first, and chunking only replaced the unwrapped function. It now runs below 16k context, where 5- and 9-position verify passes are 1.12-1.37x faster, and is on for five REAP-19B models.