Pick Apple Foundation Models 3 for runs entirely on-device on iphone - no api call, no per-token cost, works offline or the 20b sparse 'core advanced' variant lifted on-device output quality from 23% to 46% user preference in apple's own testing versus the prior generation. Pick Command A+ for cohere's newer, more efficient flagship - runs on just two h100 gpus, unusually lean for a 218b-parameter model or a 218b-total/25b-active moe released with open weights (apache 2.0) on hugging face. Choose Command A+ if you need self-hosting or data privacy; Apple Foundation Models 3 if you want a managed API.
Apple Foundation Models 3 (Apple, US) and Command A+ (Cohere, Canada) line up two different AI ecosystems against each other — a comparison that is as much about cost philosophy and openness as raw capability. Apple Foundation Models 3 is apple's on-device model family - genuinely small context and no API access, but private, offline, and free to run on every iPhone. Command A+ is cohere's newer, leaner flagship - a 218B open-weight MoE tuned for efficiency (just two GPUs to run) and 48-language support. They diverge most on context window and open vs. closed weights — each quantified below from the models' real specs.
Key differences
Cost model: Command A+ ships open weights you can self-host (hardware cost only, no per-token fee), while Apple Foundation Models 3 is API-metered at Not published. Your choice depends on whether you want zero marginal cost at the price of running infrastructure.
Context window: Command A+ holds 63× more — 256K (unconfirmed) (~384 pages) vs 4K (~6 pages). But effective recall usually fades long before the advertised ceiling, so the bigger number only helps if the model reasons over it.
Ecosystem: this is a US-vs-Canada matchup — they differ in pricing philosophy, data-residency options, and tooling ecosystems, not only benchmarks.
Specifications
Spec
Apple Foundation Models 3
Command A+
Provider
Apple (US)
Cohere (Canada)
Released
June 2026
May 19, 2026
Context window
4K (~6 pages)
256K (unconfirmed) (~384 pages)
Price (in/out)
Not published
Open weight (self-host / free)
Open weight?
No — API only
Yes — self-hostable
Modalities
text, image
text, code
SWE-Bench Verified
Not published
Not published
MRCR v2 @ 1M
Not published
Not published
Who wins what
Runs entirely on-device on iPhone - no API call, no per-token cost, works offline: Apple Foundation Models 3 — Command A+ is comparatively weak here — no official context window or per-token price published yet - both are estimated/unavailable
The 20B sparse 'Core Advanced' variant lifted on-device output quality from 23% to 46% user preference in Apple's own testing versus the prior generation: Apple Foundation Models 3 — Apple's on-device model family - genuinely small context and no API access, but private, offline, and free to run on every iPhone — and it is the newer of the two.
Deep integration with iOS apps via the Foundation Models framework, now open to third-party LLM providers too: Apple Foundation Models 3 — Apple Foundation Models 3 lists deep integration with iOS apps via the Foundation Models framework, now open to third-party LLM providers too among its strengths; Command A+ does not.
Cohere's newer, more efficient flagship - runs on just two H100 GPUs, unusually lean for a 218B-parameter model: Command A+ — Apple Foundation Models 3 is comparatively weak here — a genuinely small context window (about 4,096 tokens, shared between input and output) - far below cloud frontier models
A 218B-total/25B-active MoE released with open weights (Apache 2.0) on Hugging Face: Command A+ — Open weights make this possible at all — Apple Foundation Models 3 is API-only, so it cannot leave the vendor's servers.
Supports 48 languages, including all official EU languages: Command A+ — Cohere's newer, leaner flagship - a 218B open-weight MoE tuned for efficiency (just two GPUs to run) and 48-language support — and it carries the larger 256K (unconfirmed) context.
Largest single-prompt input: Command A+ — Its 256K (unconfirmed) window is about 63× larger than Apple Foundation Models 3's 4K, fitting roughly 384 pages in one prompt.
Which should you pick?
Someone analysing very long documents or codebases: Command A+ — Larger 256K (unconfirmed) window fits more in one prompt.
A team with data-privacy or self-hosting needs: Command A+ — Open weights let you run it on your own hardware; Apple Foundation Models 3 is API-only.
Anyone whose priority is runs entirely on-device on iphone - no api call, no per-token cost, works offline: Apple Foundation Models 3 — It is specifically built for that.
Anyone whose priority is cohere's newer, more efficient flagship - runs on just two h100 gpus, unusually lean for a 218b-parameter model: Command A+ — That is its strongest area.
An enterprise with regional data-residency rules: Apple Foundation Models 3 or Command A+ — Origin (US vs Canada) affects where data is processed and which compliance regime applies — check the provider's terms for your region.
Apple Foundation Models 3: where it fits
Apple's on-device model family - genuinely small context and no API access, but private, offline, and free to run on every iPhone. Released June 2026 by Apple, it is built for runs entirely on-device on iPhone - no API call, no per-token cost, works offline, the 20B sparse 'Core Advanced' variant lifted on-device output quality from 23% to 46% user preference in Apple's own testing versus the prior generation, deep integration with iOS apps via the Foundation Models framework, now open to third-party LLM providers too, and zero marginal cost and strong privacy - nothing leaves the device.
Its trade-offs are real: a genuinely small context window (about 4,096 tokens, shared between input and output) - far below cloud frontier models, not comparable in raw capability to frontier cloud models - built for efficiency and privacy, not benchmark leadership, and no public API pricing since it isn't sold per-token - only usable within Apple's own platforms.
Command A+: where it fits
Cohere's newer, leaner flagship - a 218B open-weight MoE tuned for efficiency (just two GPUs to run) and 48-language support. Released May 19, 2026 by Cohere, it is built for cohere's newer, more efficient flagship - runs on just two H100 GPUs, unusually lean for a 218B-parameter model, a 218B-total/25B-active MoE released with open weights (Apache 2.0) on Hugging Face, supports 48 languages, including all official EU languages, and positioned by Cohere as its 'most efficient and performant model to date'.
Its trade-offs: no official context window or per-token price published yet - both are estimated/unavailable, newer release with less independent benchmark verification than established flagships, and enterprise/multilingual focus rather than topping general reasoning leaderboards. As an open-weight model, its running cost is your own hardware rather than a per-token fee.
The bottom line for this matchup
The defining split here is open vs. closed. Command A+ gives you weights you control — self-host it, fine-tune it, keep data in-house, pay only for hardware. Apple Foundation Models 3 gives you a managed, always-updated API with no infrastructure to run. Teams with GPUs, privacy requirements, or huge volume often favour the open model; teams that want zero ops and the latest capabilities favour the closed one. Capability is close enough that this operational question, not the benchmark, usually decides it.
Frequently asked questions
Is Apple Foundation Models 3 or Command A+ better for coding?
Public SWE-Bench figures are not available for either model, so the honest test is your own repository — run an identical real bug through both. By design, Apple Foundation Models 3 leans toward runs entirely on-device on iphone - no api call, no per-token cost, works offline while Command A+ leans toward cohere's newer, more efficient flagship - runs on just two h100 gpus, unusually lean for a 218b-parameter model, and that positioning usually predicts which feels better on your codebase.
Which is cheaper, Apple Foundation Models 3 or Command A+?
Command A+ is open-weight, so self-hosting means no per-token fee (you pay for hardware instead), while Apple Foundation Models 3 is API-metered at Not published. For most teams without GPUs, the API model is cheaper to start; at very high volume, self-hosting can win.
Which has the bigger context window?
Command A+ — 256K (unconfirmed) vs 4K, about 63× larger. Useful only if the model actually reasons over the full window, which not all do.
Can I use both Apple Foundation Models 3 and Command A+ together?
Yes — a multi-model platform like LumiChats gives you Apple Foundation Models 3, Command A+ and 40+ others under one ₹69/day pass (about $1/day), so you can draft with one and cross-check with the other instead of buying two subscriptions.
Which is newer, Apple Foundation Models 3 or Command A+?
Apple Foundation Models 3 — released June 2026, about 13 days after Command A+.
Apple Foundation Models 3 vs Command A+
Apple · US | Cohere · Canada · Updated June 2026
Quick verdict
Pick Apple Foundation Models 3 for runs entirely on-device on iphone - no api call, no per-token cost, works offline or the 20b sparse 'core advanced' variant lifted on-device output quality from 23% to 46% user preference in apple's own testing versus the prior generation. Pick Command A+ for cohere's newer, more efficient flagship - runs on just two h100 gpus, unusually lean for a 218b-parameter model or a 218b-total/25b-active moe released with open weights (apache 2.0) on hugging face. Choose Command A+ if you need self-hosting or data privacy; Apple Foundation Models 3 if you want a managed API.
Apple Foundation Models 3 (Apple, US) and Command A+ (Cohere, Canada) line up two different AI ecosystems against each other — a comparison that is as much about cost philosophy and openness as raw capability. Apple Foundation Models 3 is apple's on-device model family - genuinely small context and no API access, but private, offline, and free to run on every iPhone. Command A+ is cohere's newer, leaner flagship - a 218B open-weight MoE tuned for efficiency (just two GPUs to run) and 48-language support. They diverge most on context window and open vs. closed weights — each quantified below from the models' real specs.
Key differences at a glance
▸Cost model: Command A+ ships open weights you can self-host (hardware cost only, no per-token fee), while Apple Foundation Models 3 is API-metered at Not published. Your choice depends on whether you want zero marginal cost at the price of running infrastructure.
▸Context window: Command A+ holds 63× more — 256K (unconfirmed) (~384 pages) vs 4K (~6 pages). But effective recall usually fades long before the advertised ceiling, so the bigger number only helps if the model reasons over it.
▸Ecosystem: this is a US-vs-Canada matchup — they differ in pricing philosophy, data-residency options, and tooling ecosystems, not only benchmarks.
Side-by-side specs
Spec
Apple Foundation Models 3
Command A+
Provider
Apple (US)
Cohere (Canada)
Released
June 2026
May 19, 2026
Context window
4K (~6 pages)
256K (unconfirmed) (~384 pages)
Price (in/out)
Not published
Open weight (self-host / free)
Open weight?
No — API only
Yes — self-hostable
Modalities
text, image
text, code
SWE-Bench Verified
Not published
Not published
MRCR v2 @ 1M
Not published
Not published
Who wins what
Runs entirely on-device on iPhone - no API call, no per-token cost, works offline
Apple Foundation Models 3
Command A+ is comparatively weak here — no official context window or per-token price published yet - both are estimated/unavailable
The 20B sparse 'Core Advanced' variant lifted on-device output quality from 23% to 46% user preference in Apple's own testing versus the prior generation
Apple Foundation Models 3
Apple's on-device model family - genuinely small context and no API access, but private, offline, and free to run on every iPhone — and it is the newer of the two.
Deep integration with iOS apps via the Foundation Models framework, now open to third-party LLM providers too
Apple Foundation Models 3
Apple Foundation Models 3 lists deep integration with iOS apps via the Foundation Models framework, now open to third-party LLM providers too among its strengths; Command A+ does not.
Cohere's newer, more efficient flagship - runs on just two H100 GPUs, unusually lean for a 218B-parameter model
Command A+
Apple Foundation Models 3 is comparatively weak here — a genuinely small context window (about 4,096 tokens, shared between input and output) - far below cloud frontier models
A 218B-total/25B-active MoE released with open weights (Apache 2.0) on Hugging Face
Command A+
Open weights make this possible at all — Apple Foundation Models 3 is API-only, so it cannot leave the vendor's servers.
Supports 48 languages, including all official EU languages
Command A+
Cohere's newer, leaner flagship - a 218B open-weight MoE tuned for efficiency (just two GPUs to run) and 48-language support — and it carries the larger 256K (unconfirmed) context.
Largest single-prompt input
Command A+
Its 256K (unconfirmed) window is about 63× larger than Apple Foundation Models 3's 4K, fitting roughly 384 pages in one prompt.
Which should you pick?
Someone analysing very long documents or codebases
→ Command A+
Larger 256K (unconfirmed) window fits more in one prompt.
A team with data-privacy or self-hosting needs
→ Command A+
Open weights let you run it on your own hardware; Apple Foundation Models 3 is API-only.
Anyone whose priority is runs entirely on-device on iphone - no api call, no per-token cost, works offline
→ Apple Foundation Models 3
It is specifically built for that.
Anyone whose priority is cohere's newer, more efficient flagship - runs on just two h100 gpus, unusually lean for a 218b-parameter model
→ Command A+
That is its strongest area.
An enterprise with regional data-residency rules
→ Apple Foundation Models 3 or Command A+
Origin (US vs Canada) affects where data is processed and which compliance regime applies — check the provider's terms for your region.
Apple Foundation Models 3: where it fits
Apple's on-device model family - genuinely small context and no API access, but private, offline, and free to run on every iPhone. Released June 2026 by Apple, it is built for runs entirely on-device on iPhone - no API call, no per-token cost, works offline, the 20B sparse 'Core Advanced' variant lifted on-device output quality from 23% to 46% user preference in Apple's own testing versus the prior generation, deep integration with iOS apps via the Foundation Models framework, now open to third-party LLM providers too, and zero marginal cost and strong privacy - nothing leaves the device.
Its trade-offs are real: a genuinely small context window (about 4,096 tokens, shared between input and output) - far below cloud frontier models, not comparable in raw capability to frontier cloud models - built for efficiency and privacy, not benchmark leadership, and no public API pricing since it isn't sold per-token - only usable within Apple's own platforms.
Command A+: where it fits
Cohere's newer, leaner flagship - a 218B open-weight MoE tuned for efficiency (just two GPUs to run) and 48-language support. Released May 19, 2026 by Cohere, it is built for cohere's newer, more efficient flagship - runs on just two H100 GPUs, unusually lean for a 218B-parameter model, a 218B-total/25B-active MoE released with open weights (Apache 2.0) on Hugging Face, supports 48 languages, including all official EU languages, and positioned by Cohere as its 'most efficient and performant model to date'.
Its trade-offs: no official context window or per-token price published yet - both are estimated/unavailable, newer release with less independent benchmark verification than established flagships, and enterprise/multilingual focus rather than topping general reasoning leaderboards. As an open-weight model, its running cost is your own hardware rather than a per-token fee.
The bottom line for this matchup
The defining split here is open vs. closed. Command A+ gives you weights you control — self-host it, fine-tune it, keep data in-house, pay only for hardware. Apple Foundation Models 3 gives you a managed, always-updated API with no infrastructure to run. Teams with GPUs, privacy requirements, or huge volume often favour the open model; teams that want zero ops and the latest capabilities favour the closed one. Capability is close enough that this operational question, not the benchmark, usually decides it.
Want both Apple Foundation Models 3 and Command A+ without two subscriptions? LumiChats gives you these plus 40+ models under one ₹69/day pass (about $1/day) — draft with one, cross-check with the other.
Is Apple Foundation Models 3 or Command A+ better for coding?
Public SWE-Bench figures are not available for either model, so the honest test is your own repository — run an identical real bug through both. By design, Apple Foundation Models 3 leans toward runs entirely on-device on iphone - no api call, no per-token cost, works offline while Command A+ leans toward cohere's newer, more efficient flagship - runs on just two h100 gpus, unusually lean for a 218b-parameter model, and that positioning usually predicts which feels better on your codebase.
Which is cheaper, Apple Foundation Models 3 or Command A+?
Command A+ is open-weight, so self-hosting means no per-token fee (you pay for hardware instead), while Apple Foundation Models 3 is API-metered at Not published. For most teams without GPUs, the API model is cheaper to start; at very high volume, self-hosting can win.
Which has the bigger context window?
Command A+ — 256K (unconfirmed) vs 4K, about 63× larger. Useful only if the model actually reasons over the full window, which not all do.
Can I use both Apple Foundation Models 3 and Command A+ together?
Yes — a multi-model platform like LumiChats gives you Apple Foundation Models 3, Command A+ and 40+ others under one ₹69/day pass (about $1/day), so you can draft with one and cross-check with the other instead of buying two subscriptions.
Which is newer, Apple Foundation Models 3 or Command A+?
Apple Foundation Models 3 — released June 2026, about 13 days after Command A+.
Specifications and benchmarks reflect publicly reported figures as of June 2026 and may change as providers release updates. Always verify on your own workload.